Quality Control of Carbon Fiber Structures by Acoustic Method
DOI:
https://doi.org/10.3384/wcc215.1248Keywords:
Carbon Fiber, Quality Control, Structural Dynamics, AcousticsAbstract
A method for quality control, using impact excitation and sound measurements, is investigated.
The Eigenmodes of a structure are unique if all properties, i.e. the shape, the resonance frequency, and the damping, are considered. It is, however, demanding to experimentally determine the vibration shape of the Eigenmodes, compared to getting estimates of the Eigenfrequency and the Damping.
For a specific structure it may, however, be possible to use identified peaks in a measured sound spectrum, and from them make estimates of the Eigenfrequencies and Damping, establishing enough information for quality control, even without knowing the mode shapes. The risk associated with this approach is if a match of peaks in the response spectra is found in terms of frequency and damping, but this match is in fact for different Eigenmodes (i.e. the correlation in frequency and damping is for Eigenmodes with different shapes). This may happen for one pair of Eigenmodes, but is unlikely if more Eigenfrequencies are included in the comparison.
Another risk with the used approach is if the acoustic radiation is poor from one or more of the Structural Eigenmodes, or if the excitation is sensitive to the impact (excitation) location. Further, with a single microphone used, there is a risk the sound radiation in this particular direction is low. The strategy to reduce such risks is to used multiple impact locations, and to use a non-fixed test rig for the test object.
Demonstration of the method’s applicability is made using aircraft components manufactured in Carbon Fiber Reinforced Polymer, and show promising results in terms of ability to detect variations in manufacturing quality, with a very simple test setup and short time required for the test.
Downloads
Published
Conference Proceedings Volume
Section
License
Copyright (c) 2025 Mats Gustavsson, Kent Persson, Bengt Moberg

This work is licensed under a Creative Commons Attribution 4.0 International License.