| |
|
THERMAL IMAGING NDE |
| |
|
|
| |
|
Thermal Diffusivity using Finite Pulse Widths |
| |
Thermal diffusivity measurement of a material is a very important thermal property that is considered important during design of thermally active components. This material property is also required during the non destructive evaluation for defect depth estimation through transient infrared thermography. It was shown that the scope of the flash method to determine thermal diffusivity of materials based on 1D thermal diffusion can be significantly extended by taking into account the finite pulse widths of realistic flash lamp excitations. The pulse profiles of a commercial flash system are determined and parameterized. The actual pulse shape from flash lamp system is considered and correction factors are proposed for calculating the thermal diffusivity using the flash heating method. Both experimental and numerical models (using FEMLAB® simulations) were employed to study the effect of the pulse shape on diffusivity and further validated using experiments.
A series of 1D FEM simulations are carried out to establish the notion of equivalent rectangular pulse widths. Universal response curves generated using FEM simulations and validated with experiments provide the necessary correction factors for the determination of thermal diffusivity.
It was shown that the flash method can be used even when the pulse widths are finite or when sample thicknesses are small. The universality of the extended flash method has been demonstrated through detailed FEM simulations and experiments on three different materials and a range of pulse widths encountered in practical flash systems. Further, it has been shown that the actual pulse can be represented by an equivalent rectangular pulse for simulation purposes having the advantage that it is computationally faster but without loss of accuracy. Specifically, the equivalent rectangular pulse based on the equivalent decay method using the pulse width equal to twice the width at 1/e of the maximum can be effectively used in FEM simulations instead of actual profiles. The ERP method was found to be within 3% of the results obtained using parameterized pulse shapes (obtained from measured values). The results from FEM simulations for thermal diffusivity estimates using the ERP method were verified by experiments using Aluminum 6063 samples of different thicknesses. The maximum error between the experimentally measured results and the FEM simulations was less than 2%. Appropriate material independent proportionality factors have been derived to enable thermal diffusivity estimates to be made for a wide range of material thicknesses, thermal diffusivities and pulse widths.
|
| |
|
|
|
|
FFEM simulation showing the
effect of flash duration. |
| |
|
The two methods for ERP calculation. |
|
|
Universal curve for proportionality factor. |
| |
|
|
| |
|
 |
|
| |
|
|
| |
|
Periscope Method for NDT of Pipes |
| |
A novel way using a periscope like reflector based infrared thermography to inspect local wall thinning in metal tubes, with the IR camera located outside the tube has been demonstrated.
In this work pulse thermographic techniques are applied on the tube to visualize the defects and to estimate defect depths. To simulate this wall thinning, five local wall thinning defects, each having diameter of 4 mm and depth starts from 0.5 mm with a step of 0.5 to 2.5 mm. For accessing the inner surface of the tube an aluminum reflector (a rod that was cut in a 450 plane to the axis) was used to reflect the tube’s inner diameter surface thermal emissions to an IR camera placed outside the tube at one end of the tube . Thus the temperature distribution on the inner surface of the tube is analyzed by infrared camera in this periscope approach as shown in Fig. 1. The defect depth estimation was based on the Parker’s method.
The temperature-time profiles from the FEM simulations were plotted for the Zircaloy-2 tube geometry using the FEM model. The color plots for the half model simulations (snap shots at different times from the start of the flash heating pulse) are shown in Fig. 2.
In Fig. 3, the FEM simulation results for the defect in the reflection mode is presented in a zoomed mode showing the temperature maps on the inner surface, only in the vicinity of the simulated defect.
This method may be used in certain field NDE applications where tubes need to be inspected and only the inner surface is accessible for thermographic testing. Both, the transmission and reflection modes of thermography could be successfully implemented using the periscope approach. |
|
|
|
Fig 1. Periscope approach for the estimation of temperature distribution on the inner surface of the tube |
| |
|
Fig 2. FEM simulation results for the
defect in the transmission mode |
| |
|
Fig 3. FEM simulation results for the
defect in the reflection mode |
|
|
| |
|
Reference: 1. A. Vageswar, K. Balasubramaniam, C. V. Krishnamurthy,T. Jayakumar and Baldev Raj, “Periscope Infrared Thermography for
Local Wall Thinning in Tubes”, |
| |
|
| |
|
|
|
|
|
| |
|
|
| |
|
Thermal Video Image Processing Algorithms |
| |
Thermal imaging is an advanced NDT technique based on the detection of infrared radiation, obtaining real time digital images of the Honeycomb sample and processing them. Infrared Thermography is the technique for producing an image of invisible (to human eye) infrared light emitted by objects due to their thermal condition. The most typical type of Thermograph camera resembles a typical camcorder and produces a live TV picture of heat radiation. An image produced by an infrared camera is called a Thermogram or sometimes a Thermograph. Infrared Thermograph has shown to detect the defects by monitoring the surface temperature distribution of Honeycomb. Success of automated system depends on the accurate analysis of the thermographs obtained in the infrared Thermography. Here we focus on the application of image processing technique such as edge detection, dilation, erosion, filling images for future extraction and calculating the depths of the defects using different methods.
Several thermography technologies have been developed to detect defects.
They are,
|
 |
Peak Temperature-Contrast Slope Method, (PTCSM) |
|
|
Pulse Thermography, (PT) |
|
|
Thermal Tomography (TT) |
|
|
Principal Component Thermography(PCT) |
|
|
Pulse Phase Thermography (PPT) |
|
The table below shows the relative performance of the different video image processing techniques for the defect depth calculations (in mm from top surface) for the honeycomb (AL skin-AL core) test sample with simulated defects at the interfaces. |
| |
|
Defect Numbers |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
PCM |
3.2 |
4 |
1 |
1 |
4.2 |
5 |
1.2 |
1.7 |
PT |
5 |
5 |
1 |
1 |
2 |
3 |
1 |
1 |
TT |
5.9 |
6.2 |
1.1 |
1.4 |
3.3 |
5.6 |
0.9 |
1.1 |
PCP |
3 |
4 |
1 |
1 |
4 |
4 |
1 |
1 |
|
| |
|
|
|
Experimental setups for Thermal Imaging. |
| |
|
Thermo grams at two time delays
after flash |
|
|
Temperature Contrast Plots for bottom defects (5-8) |
| |
|
|
| |
|
|
|
| |
|
Transmission mode pulsed infrared thermography |
| |
Several methods have been reported in literature for defect quantification using the pulse infrared (IR) thermography, and the peak contrast slope is one such method used to estimate the wall thickness loss (henceforth called the defect) using the reflection mode (where the IR Camera and the heating source are on the same side of the test sample). Ringermacher et al.. have proposed proportionality factor by using peak contrast slope method in the reflection mode for defect depth estimation and Sun et.al. compared this peak contrast slope method with other methods of defect depth estimation in reflection mode and found that this method is more accurate than other method. In this method, slope of the thermal contrast (∆T) at the defect region, with respect to the defect-free reference data is first determined. The time (ts) at which the peak occurs in the slope contrast plot is then related to remaining thickness of the defect (Ld) in a sample of thickness (L), in terms of the material thermal diffusivity (a) using equation (1) |
|
The proportionality factor (w) equal to 3.64 is found to be a constant independent of any material provided Ld/L  0.5 where L is the total thickness and Ld is the remaining thickness at the defect location. The well established reflection mode peak slope contrast method using a dimensionless proportionality factor was successfully extended to the transmission mode. The proportionality factor was established to be equal to 0.9 independent of any material using an analytical model and validated with experiments for Ld/L £ 0.5. For y>0.5, a graph relating the proportionality factor and y is provided to evaluate defect depths. The predictions using the analytical expression were corroborated with FEM simulations and validated with experiments on Steel (to within 4%). The transmission mode is shown to provide greater accuracy over the reflection mode in defect depth estimation.
|
|
|
|
Experimental Setup |
| |
|
Thermo gram of wall thickness loss |
| |
|
The universal plot for the TT mode for diffusivity measurement. |
|
|
| |
|
| |
|
|
|
|
