Researchers Improve Capability to Detect Cryogenic Tank Damage Published July 18, 2007 By Materials and Manufacturing Directorate AFRL/ML Wright-Patterson Air Force Base, Ohio -- Researchers from AFRL, the University of Dayton Research Institute, and Purdue University's School of Mechanical Engineering made significant progress in developing an improved capability for monitoring the structural health of cryogenic tanks used for space missions. The researchers conducted highresolution nondestructive evaluation (NDE) and structural health monitoring (SHM) experiments using elastic wave propagation (in the form of surface and guided waves) and were able to quantify microstructure and mechanical damage in friction-stir-welded aluminum-lithium alloys, which are prominent candidates for cryogenic tank construction. The research is not only essential to ensuring the reusability of cryogenic tanks across hundreds of missions, but is applicable to other types of unitized structures as well. The improved capacity to detect incipient damage in cryogenic tanks enhances safety, speeds turnaround time, and reduces operating and material costs. Therefore, AFRL's NDE and SHM research benefits the Air Force, future space programs, and industry alike. Researchers examined the dynamic behavior of the alloy plates and determined nodal and antinodal points. They then applied both high-frequency (elastic wave propagation) and low-frequency (vibration) acoustic waves to interrogate the damage in the plates. The team initiated high frequencies using piezoelectric patches/transducers and low frequencies using a dynamic shaker system. The researchers also used burst and swept waveforms for the high frequencies. Next, the team simulated different forms of damage, using either localized temperature gradients or mass placement along and away from the weld. The team's effort also involved damage studies conducted by introducing cracks in the plates at various locations. Sensors placed at different locations enabled researchers to understand sensor effectiveness in capturing the damage data. To analyze the time-domain data obtained from the sensors, researchers used signal processing methods such as conversion to frequency-domain data, time-frequency analysis, and harmonic wavelet analysis (with each wavelet level corresponding to an octave band of traveling plate wave modes). Additionally, they assessed statistically significant features for damage detection and identified damage locations using beamforming methods. The researchers' numerous experiments were able to show that the vibroacoustic (combined low- and high-frequency) method is an effective and appropriate way to interrogate damage in both welded and nonwelded test plates.