Projects are carried out each spring by groups of students enrolled in a class called ME597A ‘‘Practical Experiences in Vibration’’ at Purdue University in the School of Mechanical Engineering. As the first installment of a three part series, the experiments and results of a student team’s project are presented covering the analysis of a C-130J aircraft fuselage search-and-rescue window design modification for the U.S. Coast Guard using modal impact and operational structural acoustic testing.
This course, and others like it at institutions across the country, recognizes that experimental vibrations and acoustics are best learned if students are required to work in groups to conceptualize, design, conduct, and analyze their own experiments. Multiple offerings of this course at Purdue have demonstrated that students retain more knowledge of measurement systems, signal processing, frequency response estimation, modal analysis, and other topics in experimental dynamics if they experience the frustrations and revelations of testing first hand. Students sometimes find testing tedious; however, they always come away from the course with a solid understanding of the basics in structural dynamic testing: What is modal impact testing? How can it be used to obtain information about the free vibration response of a structure? What signal processing issues can cause problems in the analysis of data?
In this experimental vibrations course, the instructor serves primarily as a ‘‘measurement coach’’ by providing students with quick tutorials and suggestions at various points throughout the semester. By working together, students usually troubleshoot and solve their own measurement challenges. In fact, different groups often consult for one another during the semester leading to a highly collaborative learning environment.
In addition to benefiting students, this type of experiential and cooperative learning approach also benefits industry. In fact, the course at Purdue is sponsored by notable industry partners (including Arvin Meritor, General Motors, Lord Corporation, The Modal Shop, PCB Piezotronics, among others) who have provided test structures, measurement instrumentation and expertise in-kind. Surveys of these sponsors from both industry and federal laboratories with an interest in structural dynamics indicate that by allowing students to conceptualize and design their own experiments, open-ended lab experiences provide students with the training necessary to hit the ground running when they enter the workplace. For example, students are exposed to the latest sensing, data acquisition, and data analysis (software) technologies during the semester-long course at Purdue. VXI Technologies’ 1432 data acquisition systems with Multi-Reference Impact Testing (MRIT) and X-MODAL interfaces, which were created by the University of Cincinnati Structural Dynamics Research Laboratory, are used by every student in the course. This exposure to multi-channel dynamic measurement systems helps to demystify experimental techniques so the student can quickly learn whichever system they are expected to use in industry. Surveys and hiring histories associated with programs where hands-on, open-ended labs are offered demonstrate that employers are more likely to hire undergraduate and graduate students who have participated in such programs.
Fig. 1: C-130J U.S. Coast Guard aircraft for search & rescue missions
PROJECT C-130J WINDOW MODIFICATION
The U.S. Coast Guard (USCG) is considering adding observation windows to the sides of the C-130J fuselage (see Fig.1) in order to use it for search and rescue missions. The USCG is concerned with the potential noise and vibration problems that could arise as a result of this structural modification especially given the large volumes of air displaced by six propellers on each of the four engines. The C-130J already has more significant vibration and noise levels than the C-130H, which has only four blades per engine. The aim of this project was to compare the vibration and sound transmission characteristics of two different fuselage sample sections, one with a window and one without a window. The test results were analyzed to determine if the noise and vibration changes resulting from the window structural modification will be tolerable by the crew for long duration search and rescue flights.
In the initial phases of the project, students were required to develop an experimental methodology for assessing the relative noise and vibration characteristics of the fuselage with and without a window. The principal challenge in this project was that the C-130J could not be made available for testing. Therefore, the students chose to work with another department at Purdue, the Department of Aviation Technology, to construct the specimens they required for testing. The two groups of students chose to construct two mock fuselage sections so that tests could be conducted on each section. There were many lessons learned by this group of engineering students through working with the technology students and vice versa. The technology students had all of the knowledge necessary to construct the sections; however, they required data on specimen geometry, desired method for supporting the specimens, etc. The technology students in turn learned about free and forced response, modal impacts testing, and acoustic testing.
Lesson Learned #1
Both groups of students learned that experimental structural dynamic projects are rarely carried out by one engineer or group in a vacuum. Multiple groups of engineers and technicians almost always coordinate on tests of this nature. In order to coordinate these tests properly, all assumptions must be defined beforehand so that specifications can be exchanged among the groups. For these tests, assumptions were defined relating to the effects of differences in the stringers and doublers of the fuselage test sections; specimen thickness (0.071 (1.80 mm) in was used instead of 0.041 (1.04 mm) in because of availability); lack of tempering on fuselage material; etc. Most of these assumptions led to the conclusion that the mock fuselage natural frequencies would be lower than expected in the real aircraft. Therefore, when the fuselage sections respond to the simulated operational vibration and sound spectra, larger responses at lower frequencies were anticipated from the outset of the project.
The mock fuselage sections (panels) were manufactured by students from the Department of Aviation Technology. The students used 2017 (skin) and 7075 (frame) aluminum alloys and geometries that are characteristic of a real C-130J fuselage. Three thousand MS20470AD4 rivets were used to construct each panel. Each panel was approximately 80 inches (2 m) tall by 48 inches (1.2 m) wide by 0.071 in (1.80 m) thick. One of the two sections was created with the planned observation window in the middle of it. The dimensions of the window were 30 inches (0.76 m) tall by 22 inches (0.56 m) wide and made of double-layered Plexiglas. Figure 2 shows the stringer configuration used in the windowless panel construction.
Controlled vibration and sound experiments were performed on two small sections of a portion of the fuselage of a C-130J where the window would potentially be added. As the engineering students began to develop their experimental strategy, they soon learned the difference between the ‘‘free response’’ and ‘‘forced response’’ of a structural system. They also learned that different experimental techniques are required to identify the free and forced response characteristics.
Fig. 2: Windowless panel stringer configuration used in mock fuselage section
Lesson Learned #2
Students developed an appreciation for free and forced response testing in structural dynamics. The free response characteristics, which included the natural frequencies and modal deflection shapes, of the fuselage sections could be obtained using modal impact testing. Modal impacts provide a broadband excitation to the specimens in order to excite a full range of natural frequencies and vibration shapes. On the other hand, the forced response characteristics are the product of the fuselage’s preference for responding at certain natural frequencies and the operational excitation. One component of the operating excitation flows through the airframe structure (structure borne component) and another component flows through the fuselage (airborne component). To identify the airborne component, students would be required to simulate the non-contact acoustic pressure from the propellers using an electromechanical loudspeaker.
Fig. 3: Windowless (left) and windowed (right) fuselage sections shown suspended on nylon ropes and instrumented with accelerometers
Fig. 4: Modal impact degrees of freedom utilized in tests on fuselage sections
To compare the vibration characteristics of the two panels, modal impact testing was performed using five single-axis ICP® accelerometers (PCB Piezotronics model 333B32 100 mV/g) as references, oriented transverse to the fuselage surface and attached with mounting wax. Impacting 77 locations using a Modally Tuned ICP impact hammer (PCB Piezotronics model 086C03 10 mV/ lb), a total of 385 frequency response functions (FRF) measurements were made. Figure 3 shows the windowless (left) and windowed (right) panels hung for testing on nylon ropes to simulate a free-free boundary condition in the transverse direction. FRF data was acquired using an 800 Hz frequency span with 3200 spectral lines for a 0.25 Hz frequency resolution. The MRIT (Multi-Reference Impact Testing) software package was used to acquire this data. Figure 4 shows the measurement degrees of freedom (inputs and outputs) used in these tests. During testing, students observed that certain points on the panel were very compliant leading to a large roll off in the input auto power spectrum at around 400 Hz. Although tests were continued and the data was analyzed despite the large changes in excitation bandwidth from point to point, those points were eventually moved inward on the panel to avoid any reduction in excitation bandwidth.
Lesson Learned #3
One important lesson learned during modal impact testing was that the input power spectra should be monitored closely throughout a test on a large structural component with complicated characteristics. In the case of the fuselage section with stringers and doublers, impacts on certain positions led to very low bandwidths of excitation. When students extracted the modal frequencies and mode shapes, it was apparent that the modal properties at certain impact points with low compliance were in error because the fuselage was not properly stimulated at those frequencies. When conducting modal impact testing, modal impacts are supposed to resemble true impulses, which excite all natural frequencies equally; however, real impacts are not impulses and can degrade at certain points across a large structural component.
Fig. 5: Modal deflection shapes at 41 Hz and 63 Hz for windowed panel
Fig. 6: Acceleration frequency response function magnitudes for impact location at point #39 on windowless and windowed panels
The X-MODAL software package was then used to extract the modal parameters (frequencies and mode shapes). Students learned about the benefits of single-reference vs. multi-reference modal parameter estimation. Figure 5 shows two example mode shapes for the windowed panel. The panel acts as a monopole and dipole sound radiator at 41 Hz and 63 Hz, respectively. In order to compare the vibration response of the two fuselage sections, the FRFs in Fig. 6 were compared for an impact applied at point #39 on the windowless (blue) and windowed (red) panel with responses at points #16, 41, 46, and 59, which are distributed near the panel center where the window is located. Note that the presence of the window causes a reduction in the forced vibration response above 200 Hz but an increase in response at several resonant frequencies below 200 Hz.
Lesson Learned #4
The important lesson learned regarding the changes in vibration from the windowed specimen to the windowless specimen is that when vibrations increase in one frequency range, vibrations nearly always decrease in another frequency range. This general rule in structural dynamics is one of the challenges in designing for noise and vibration performance; design modifications must be chosen so that reductions in FRFs in one frequency range do not lead to amplifications in FRFs for other frequency ranges where the operational excitation is large.
If the operational spectrum of the C-130J with its four, six bladed propellers is considered, it can be concluded that low frequency vibration and noise are likely to worsen leading to body shake but high frequency vibration and noise are likely to improve leading to reduced amplitudes of higher frequency noise tones. As mentioned previously, this kind of design trade-off in structural dynamic systems is common whereby a design causes a decrease in one frequency range and an increase in another.
Fig. 7: Acceleration frequency response function for an acoustic excitation on windowless and windowed panels
To substantiate these results for noise transmission, a loudspeaker was used to excite each fuselage section and accelerometers were used to measure the panel structural vibrations resulting from a broadband random acoustic load with frequency content from 0.1–1000 Hz. The assumption made was that if the accelerations increased on the inner surface of the fuselage section, then the noise would increase as well. Note that the radiation efficiencies of the fuselage aluminum and Plexiglas materials were not taken into account. Figure 7 shows the results of this comparison for the windowless and windowed panels. These kinds of assumptions are usually necessary in student projects performed over a single semester. The first characteristic that students noticed in these plots was the roll off of the responses at all response measurement degrees of freedom.
Lesson Learned #5
The relevant lesson learned here is that actuators are imperfect devices with dynamics of their own. Acceleration measurements made on the cone of the loudspeaker indicated that the roll off in response was due to the roll off in the loudspeaker’s excitation bandwidth, not the filtering properties of the fuselage section. However, in this experiment, a side by side comparison is being made between two fuselage sections, so the roll off in excitation bandwidth did not pose a serious problem because both fuselage sections were excited by the same sound pressure spectrum.
Figure 7 indicates that the windowed panel clearly has less vibration response to the structural acoustic excitation above 200 Hz but slightly higher resonant response below 200 Hz. The trade offs involving increases and decreases in noise response in this case are again evident as they were in the vibration response.
A potential window design modification to a fuselage was examined to assess the affects on noise and vibration to crew inside the aircraft. Modal impact testing was first used to compare the structural vibration characteristics of two fuselage sections with and without windows. Students learned valuable lessons about how to define their assumptions so that specifications could be exchanged with other group members. Students also learned the differences between free and forced response. Because the fuselage sections were comprised of thin panels separated by stiff stringers, students also observed that the compliance to modal impacts varied substantially from point to point on the test specimodal impact hammers, piezo drivers, and other actuation devices, have dynamics of their own leading to important changes in the forced response of structural dynamic systems to those excitations.
Students showed that the forced vibration response amplitudes to modal impacts above 200 Hz were less with a window in the fuselage but greater below 200 Hz. Similarly, it was shown that forced vibration response to acoustic excitations above 200 Hz dropped and those below 200 Hz increased slightly. These trade-offs observed by students in this project are common in structural design where vibration or noise is amplified in one frequency range and attenuated in a different range by a certain design modification.