Abradable aluminum-silicon/polyester (AlSi-PE) coatings are sprayed onto the inner surfaces of aircraft engine casings to increase their overall efficiency by reducing clearance while limiting damage caused by potential interactions between the blade tips and the casing. To accurately simulate the abrasive behavior of such materials, numerical models require experimental data obtained under representative conditions. Although constitutive laws for compression loading at high strain rates were published recently, the shear behavior of abradable coatings still remains largely unexplored and poorly understood. This study presents preliminary results on the shear behavior of AlSi-PE material thermally sprayed for a specific orientation subjected to high strain rates (3500 s ^-1 and 8000 s ^-1 ) obtained using a Split Hopkinson Pressure Bar (SHPB) device incorporating a Tandem Momentum Trap (TMT), thus subjecting the sample to a single load. Post-mortem and in-situ observations were used to validate the plastic deformation zone and show the fracture behavior of AlSi-PE. The similarity between the two sets of data indicates that no pronounced strain-rate effect was observed within the accuracy and limited sample size of the present study, but post-mortem observations reveal complex fracture mechanisms related to the shear load. The in-situ experiment confirms that the sample was subjected to a predominant shear load, with apparent minor tension and compression solicitation due to the specimen geometry. The proposed study provides a first approach to shear characterization and offers significant observations on the fracture behavior of abradable AlSi-PE in dynamic conditions.
In conventional digital image correlation (DIC), full-field displacement and strain are obtained by correlating images with a reference image as the system evolves forward in time. However, in several processes, a reference image is not available (e.g., additive manufacturing), or the inherent causality of the process introduces complications and additions to the field of view (e.g., formation of new cracks within the field of view in fracture analyses), limiting the applicability of conventional DIC. This work evaluates the feasibility of performing DIC in reverse time to estimate the evolution of forward-time strain fields. A theoretical framework based on incremental DIC is first developed and discussed. The methodology is then assessed using quasi-static tensile tests on AA6061 specimens characterized via DIC. Errors between forward-time measurements and remapped reverse-time results are quantified at both pointwise and field-averaged levels. The approach is further demonstrated through a case study on concrete additive manufacturing, where strain field evolution is evaluated using the proposed method. The results demonstrate that reverse-time DIC can accurately reconstruct displacement and strain fields, making it particularly useful for processes where a reference image is unavailable. In the context of concrete additive manufacturing, the method enables the measurement of residual strains arising from complex, transient chemo-mechanical deformations during the process. Reverse-time DIC provides a viable framework for direct experimental monitoring of in-process deformation fields, with strong potential for application in additive manufacturing and other scenarios where conventional DIC is limited.
Blade tip timing (BTT) provides a non-contact and online means of measuring rotating blade vibration, but the resulting under-sampled signals require reliable identification algorithms. A revised compressed-sensing BTT method with total phase correction (RCS) was previously developed by the authors and validated through numerical and laboratory-bench studies; however, its performance under realistic compressor-rig conditions has not yet been experimentally established. This study aims to provide the preliminary compressor-rig validation of the previously developed RCS method and to systematically evaluate its practical performance under different measurement conditions. BTT measurements were acquired on a single-rotor compressor test rig during a slow run-up process. RCS was compared with the traditional compressed sensing method (TCS) under different sensor numbers, randomly selected sensor arrangements, and different numbers of sampling cycles. A high-density 36-sensor spectral analysis result was used as the reference for relative comparison. For three representative sensor arrangements, RCS was further compared with the established circumferential Fourier fitting (CFF) and Bayesian linear regression (BLR) methods. In addition, an ℓ_1/2 -norm-based iterative reweighted ℓ_1 algorithm was applied to both the RCS and TCS sensing models to evaluate the influence of a stronger sparsity-promoting solver. RCS accurately identified the dominant engine order and vibration amplitude of the compressor blade. For all sensor numbers considered, its average relative amplitude error remained below 5
Experimental Modal Analysis (EMA) of thin and lightweight rotating structures is challenging due to the difficulty of applying controlled non-contact excitation. Existing non-contact excitation methods often suffer from limited spatial precision. Seek a set of design and measurement parameters of an acoustic non-contact excitation device for thin and lightweight structures, such that the effective excitation area is comparable to that of impact hammers. Parametric Array Loudspeaker (PAL) is utilized to generate amplitude-modulated ultrasonic waves, producing a highly-directive acoustic field. An exponential horn is placed in front of the PAL to enhance acoustic focus. Directivity is evaluated by an area scan through the area. The validated system is duplicated and two-point excitation EMA of real artefact is carried out. The proposed system, consisting of a PAL and an exponential horn, achieves an effective excitation area of around 216 mm^2 . The system has demonstrated its ability in supressing the responses of selected modes. The proposed non-contact excitation system enables the excitation location with precision comparable to impact hammers. Its feasibility as an effective excitation approach for multiple-point excitation of thin and lightweight structures is demonstrated.
Experimental modal analysis with laser Doppler vibrometry is usually limited to the identification of mono-dimensional mode shapes, typically in the out-of-plane direction. This simplification restricts the understanding of the true three-dimensional deformation of a structure. The objective of this work is to develop an experimental–computational framework for the reconstruction of full three-dimensional mode shapes using a single scanning laser Doppler vibrometer and to establish a consistent procedure for comparison with numerical mode shapes expressed in different coordinate systems. The methodology combines frequency response function measurements obtained from three independent orientations of the tested component while maintaining a fixed laser head position. A reconstruction algorithm based on projection relations and rotation matrices is employed to recover the complete three-dimensional vibration shape, while an additional least-squares-based projection algorithm is introduced to project numerical mode shapes into the experimental reference frame. The methodology is experimentally validated on an air-compressor collector using a Polytec PSV-500 scanning laser Doppler vibrometer. The reconstructed experimental three-dimensional mode shapes showed strong agreement with numerical predictions for the first three vibration modes, with Modal Assurance Criterion values of 0.93, 0.94, and 0.85, confirming strong modal correspondence. The proposed framework provides a cost-effective and flexible alternative to multi-head vibrometer systems for three-dimensional experimental modal analysis and enables reliable reconstruction of three-dimensional mode shapes together with consistent comparison with numerical models.
