Search

Real time damage detection and characterization during load testing are valuable for understanding composite performance. The early detection and characterization of damage progression are critical for understanding failure modes. When a single stringer panel is subjected to significant quasi-static loading, the deformation of the panel results in damage such as subsurface delaminations. Surface temperature signatures and acoustic emissions enable real time detection of damage formation. Surface temperature signatures are useful for real time imaging of the damage. The time evolution of these signatures is complex. There is an initial, almost instantaneous component with a timing that is independent of the depth of the damage. This timing is based on the arrival of the acoustic emission signal at a rate orders of magnitude faster than the thermal diffusion. This signature decays with time. A second component increases with time, as the heat from damage formation diffuses to the surface. To better characterize the damage, computational simulations of the thermal signatures are performed. The simulations are shown to assist in separation of the two components and provide an estimation of the damage depth. [ABSTRACT FROM AUTHOR]

Passive thermography is used for real time damage detection during quasi-static load testing of a composite single stringer hat stiffened panel. When significant subsurface damage is detected the composite panel is unloaded to enable a more precise determination of damage extent by X-ray CT or ultrasound inspections. This enables characterization of damage growth through-the-thickness and provides critical information to validate structural designs and damage progression models. A significant advantage of passive thermography is precise, real time location and imaging of damage thus making structural testing more efficient. Passive thermography data were processed to enhance defect contrast for comparison to other NDE techniques. In particular, passive thermography data have shown good correlation between the real time thermal indications and the detected acoustic emission events. A two-dimensional thermal simulation is compared to the thermal data for damage depth estimation. Lastly, passive thermography data are compared to ultrasonic through-the-thickness measurements. [ABSTRACT FROM AUTHOR]

Previous passive thermography efforts to monitor the progression of damage during cyclic fatigue testing have focused on the heat generated at the cyclic load frequency. A closer examination of the acquired data has shown that at some locations on a test panel, the passive thermal response has significant content at twice the cyclic frequency. The locations of responses, with the largest harmonic contributions, are typically at the edge of subsurface delaminations based on ultrasonic scans of the panel. A physics-based model is presented that describes how a harmonic signal is expected if there is a friction based heat source. From the model, the expected phase of the heat source is calculated. The source is inserted in one-dimensional and three-dimensional simulations for heat diffusion to calculate the surface response for a subsurface friction source. The phase of the harmonic is found to agree reasonably well with a source at a delamination depth in the same location on the test panel as determined from ultrasonic scans of the damage. [ABSTRACT FROM AUTHOR]

Passive thermography has been shown to be an effective method for in situ and real time nondestructive evaluation (NDE) to measure damage growth in a composite structure during cyclic loading. The heat generation by subsurface flaws results in a measurable thermal profile at the surface. To better understand these sources, the quadrupole method is used to simulate the resultant temperature profile for a subsurface heat source. The simulations are compared to finite difference and finite element simulations of the same sources and experimental data acquired during cyclic loading of a composite specimen. [ABSTRACT FROM AUTHOR]

Passive thermography and acoustic emission data were obtained on a three stringer panel during periodic fatigue loading. The acoustic emission data were mapped onto thermal data, revealing the cluster of acoustic emission event locations around the thermal signatures of interest. By combining both techniques, progression of damage growth is confirmed and areas of failure are identified. Furthermore, sudden changes in thermally measured damage growth related to a previously measured higher energy acoustic emission event are studied to determine damage depth. A thermal model with a periodic flux heat source is presented to determine the relationship between the damage depth and thermal response. The model results are compared to the measured data. Lastly, the practical application and limitations of this technique are discussed. [ABSTRACT FROM AUTHOR]

Passive thermography and acoustic emission data were obtained on a three stringer panel during periodic fatigue loading. The acoustic emission data were mapped onto thermal data, revealing the cluster of acoustic emission event locations around the thermal signatures of interest. By combining both techniques, progression of damage growth is confirmed and areas of failure are identified. Furthermore, sudden changes in thermally measured damage growth related to a previously measured higher energy acoustic emission event are studied to determine damage depth. A thermal model with a periodic flux heat source is presented to determine the relationship between the damage depth and thermal response. The model results are compared to the measured data. Lastly, the practical application and limitations of this technique are discussed. [ABSTRACT FROM AUTHOR]

Numerical simulations of thermographic responses in composite materials have been useful for evaluating and optimizing thermographic analysis techniques. Numerical solutions are particularly beneficial for thermographic techniques, since the fabrication of specimens with realistic flaws is difficult. A quadrupole method for performing the simulations in two dimensions is presented. The results are compared to a finite element simulation of the same geometry. The technique is shown to be in good agreement with a finite element simulation of the same geometry, however, it requires about one hundredth of the computational time. [ABSTRACT FROM AUTHOR]

Conventional methods for reducing the pulsed thermographic responses of delaminations tend to overestimate the size of the flaw. Since the heat diffuses in the plane parallel to the surface, the resulting temperature profile over the flaw is larger than the flaw. A variational method is presented for reducing the thermographic data to produce an estimated size for the flaw that is much closer to the true size of the flaw. The size is determined from the spatial thermal response of the exterior surface above the flaw and a constraint on the length of the contour surrounding the flaw. The technique is applied to experimental data acquired on a flat bottom hole composite specimen. [ABSTRACT FROM AUTHOR]

A method for determining the depth of delaminations in composite specimens is presented. The method is based on a one-dimensional model for a composite with a delamination that is represented as a contact resistance between the upper and lower regions of the specimen. To estimate the depth and contact resistance of a delamination in an efficient method, a method is used that makes use of a eigenvector representation of the measured and theoretical data. The technique is shown to give estimates of manufactured delamination depths that are in good agreement with the specified depths. © 2003 American Institute of Physics [ABSTRACT FROM AUTHOR]

A measurement configuration is discussed where shutters are placed in front of the imager and the flash lamps. Proper timing for the opening and closing of the shutters eliminates a direct interaction between the imager and flash lamps for a single sided measurement, enabling accurate measurement of the thermal response of a specimen. The time dependence of the thermal response is analyzed to produce an image of the effective diffusivity of the specimen. [ABSTRACT FROM AUTHOR]

Quartz lamps are a common excitation source for thermal inspection systems. In contrast to flash lamps which produce short duration heat flux (impulse heating), quartz lamps produce a long pulse of heat flux (pulse heating). A method is described to measure the input heat flux from the quartz lamp incident light using a high sample rate noncontact temperature measurement system. The method includes an analysis technique for separating the convection loss from the quartz lamp heat flux in the characterization. Measuring the input heat flux enables separation of the lamp’s response from the structure’s response to yield accurate quantitative characterization. To demonstrate the utility, measurements were performed on specimens with variable thickness. The thickness is estimated by fitting the specimen’s thermal responses with theoretical expressions. Systematic errors in thermal estimation of the thickness are demonstrated for theoretical expressions when the heat flux profiles are not taken into account. Finally, a comparison is made of two correction techniques. The first technique corrects for the quartz lamp temporal heat flux profile by introducing a time shift in the analytical solution. The second technique uses the convection loss corrected temporal heat flux profile in a time domain convolution. © 2000 American Institute of Physics. [ABSTRACT FROM AUTHOR]