Kaci J. Lemler - Unmanned Aircraft Systems Engineering Laboratory, School of Engineering and Mines University of North Dakota
William H. Semke - Unmanned Aircraft Systems Engineering Laboratory, School of Engineering and Mines University of North Dakota
The Unmanned Aircraft Systems Engineering (UASE) laboratory team operates a Bruce Tharpe Engineering (BTE) Super Hauler UAV. The UASE team develops payloads up to 30 lb (14 kg) for intelligence, surveillance, and reconnaissance (ISR) missions, as well as airborne sense and avoid (ABSAA) systems. The Super Hauler is 10 ft long (3 m) and has a 12 ft (3.6 m) wingspan and a dry weight of 48 pounds (22 kg). The aircraft’s behavior in its original configuration is well understood, but adding wing payload pods for future missions will change the aircraft’s structural and flight behavior.
The Super Hauler is custom designed to provide a large, unobstructed payload bay to mount multiple payloads for flight testing. This capability meets the needs of the previous payloads tested, but adding wing payload pods is necessary to conduct new missions and evaluate payloads for radar-related work. The wing payload pods will most often be used in pairs to equalize the wing loading and provide a symmetric load on the airframe. We will run the electrical and power connectors and wiring inside of the wing so they are out of the airflow. In this manner, the antenna systems inside the wing pods have an uninterrupted front-facing field of view.
Our goal is to obtain a full modal analysis of our UAV to better understand its flight characteristics and the effect of the wing pods, and to help determine appropriate flight conditions for operation. This sort of testing has been done before on many different kinds of aircrafts, but we hope that our work will assist others that are testing small UAVs in a similar manner.
Experimental Modal Analysis
For the analysis, we used ModalVIEW software, which was developed by ABSignal using NI LabVIEW software and is available on the LabVIEW Tools Network. We chose this software because of its ease of use and perfect fit for the type of testing we are doing. We used ModalVIEW and NI CompactDAQ hardware to create a plug-and-play system. ModalVIEW was the only modal analysis program we found that had a easy to understand interface for every step of the vibration testing process, from connecting the sensors to animating the structure with the mode shapes that were found.
We used experimental modal analysis to obtain structure modal parameters by measuring and analyzing the dynamic response of the structure when excited by a stimulus. ModalVIEW supports a variety of multiple-degree-of-freedom (MDOF) global fitting analysis methods and multiple-input, multiple-output (MIMO) polyreference experimental methods using impact hammers or shakers. It also has an animation tool that we used to visualize the aircraft’s mode shape.
We isolated the Super Hauler for analysis by placing it on bungee cords in a test rig so all the wheels were 1.25 in. off the ground, which simulates a free-free boundary condition for modal testing. We mounted a 5 lb weight on each wing to simulate the wing pods. To perform the analysis, the BTE Super Hauler was first instrumented with accelerometers in key locations. We mounted a mixture of PCB Piezotronics triaxial and uniaxial accelerometers directly on the aircraft’s outer skin. We connected the accelerometers to an NI cDAQ-9178 chassis containing eight NI 9234 dynamic signal acquisition (DSA) modules. The Super Hauler was excited with a PCB Piezotronics 086C03 impact hammer.
The NI CompactDAQ front end interfaced seamlessly with the ModalVIEW software for the analysis. The DAQ channels connected to the appropriate nodes and measurement directions with measurements set to trigger with excitation from the impact hammer. We performed six impact tests both with and without the simulated wing pods attached. Each of the tests consisted of impacting the Super Hauler at one of the nodes, measuring the frequency response, then obtaining the natural frequencies from the frequency response function (FRF) and animating the line model with the corresponding mode shapes. It is speculated that the first natural frequency without the pods doesn't have a match in the set with the pods because there isn't as much structural coupling between the wings and the fuselage.
In the mode shapes observed, the natural frequencies all shifted dramatically down with the addition of the wing pods. This was expected since a substantial amount of weight was added. The wing pods account for 20 percent of the total weight, therefore, they have a fairly dramatic influence on the structural response. Although significant changes in frequencies were observed, none of the fundamental modes shifted out of the operational frequency range of the aircraft without the pods installed. Therefore, it is expected that changes in flight dynamics will not be a significant factor in flight performance.
Kaci J. Lemler
Unmanned Aircraft Systems Engineering Laboratory, School of Engineering and Mines University of North Dakota