Pulse Compression Approach for Frequency Modulated Thermal Wave Imaging Based Subsurface Defect Analysis

Size: px

Start display at page:

Download "Pulse Compression Approach for Frequency Modulated Thermal Wave Imaging Based Subsurface Defect Analysis"

Tracey Osborne
5 years ago
Views:

1 More Info at Open Access Database Pulse Compression Approach for Frequency Modulated Thermal Wave Imaging Based Subsurface Defect Analysis Aparna Akula 1, 2,a, Ravibabu Mulaveesala 3, Suman Tewary 1, 2, Ripul Ghosh 1, 2, Satish Kumar 2 and H K Sardana 1,2 1 Academy of Scientific and Innovative Research, Rafi Marg, New Delhi , India 2 CSIR-Central Scientific Instruments Organisation, Chandigarh , India 3 Indian Institute of Technology Ropar, Nangal Road, Rupnagar, Panjab , India a aparna.akula@csio.res.in Keywords: Active Infrared Thermography, Pulse Compression, Hilbert Transform, Frequency modulated thermal wave imaging Abstract. Active Infrared Thermography is a non-contact and nondestructive inspection method used to expose the surface or subsurface abnormalities in a test sample. The dynamic temperature response of the defective region is unlike the non-defective region which is the main concept behind IR thermography. Variants of active IR techniques such as pulsed thermography, lock-in thermography and pulsed phase thermography have been employed by researchers in the earlier literature. This paper highlights the successful application of recently proposed frequency modulated thermal wave imaging for defect detection. The phase and cross-correlation image sequences are generated from the captured IR image sequence using the proposed Hilbert transform based pulse compression approach. Experiments were conducted using an active thermography system comprising of mid wave infrared range (3-5 µm) IR camera, two halogen light sources with source control via a signal generator. A sine-modulation heating, varied from 0.01 Hz 0.1 Hz for duration of 100 seconds was submitted to the mild steel specimen which causes a similar frequency modulated surface heating over the specimen. The images of resolution 320 x 256 pixels were acquired at a frame rate of 20 Hz after the heating is removed. The specimen contains six numbers of circular flat bottom holes, at different depths from the specimen top surface. It was observed that the proposed method detects defects at different depths and also, filtering of images to remove high frequency noise enhances the defect detection capabilities. This paper presents in detail the experimental methodology, results of pulse compression approach for investigation of sub-surface defects like voids. Introduction Infrared thermography is widely used for non-destructive testing (NDT) applications [1]. Thermographic methods monitor the heat flow over the structure. The presence of defect disrupts the normal heat flow and causes a temperature gradient. This temperature gradient is measured by using an infrared camera for defect detection. Infrared thermography is broadly classified into two categories, active and passive. In passive thermography, defects are visible because of the natural temperature difference between the defective and non-defective regions. However, the thermal gradient is not so prominent in case of deeper and smaller defects and are not visible on the surface in case of passive thermography approach. This is overcomed in the case of active thermography where the specimen is excited by external heat stimulus and the thermal response to this stimulus on the specimen surface is monitored by an infrared camera. Due to its advantage of probing deeper defects, active thermography has become popular in the field of NDT [2].

2 Researchers are working extensively and introducing new active thermography methods. The most popular thermal NDT methods are pulse thermography [3], stepped thermography [4], lock-in thermography [5] and pulsed phase thermography [6]. Pulsed thermography is simple and widely used method, where the specimen surface is subjected to short duration high peak power heat burst and analysis is carried out during the cooling process. High peak powers are demanded for deeper subsurface analysis added to it non-uniform heating and non-uniform emissivity, limits its detection capabilities. This drawback is not existent in lock-in thermography which uses a periodic sinusoidal thermal excitation at very low peak powers as compared to the pulse thermography approach and the use of phase image for analysis instead of magnitude makes it less sensitive to thermal noise. However, selection of suitable frequency to resolve defects at different depths limits the applicability of this method. Pulsed phase thermography combines the advantages of both the approaches; it makes use of the capability of the phase method for deeper subsurface analysis, with a simple pulse excitation. The recently introduced frequency-modulated thermal wave imaging (FMTWI) approach [7] overcomes the limitations of the previous techniques by sweeping a suitable band of frequencies into the specimen in a single experimentation cycle with relatively low peak power sources; coupled with advanced signal processing methods FMTWI allows detection of deeper defects. This paper demonstrates the Hilbert transform based pulse compression approach for FMTWI for characterizing the mild steel sample for subsurface defect detection. Frequency-modulated thermal wave imaging In FMTWI heat flux modulated by a suitable band of frequencies of equal energy are introduced into specimen to probe for detection of subsurface abnormalities at different depths in a single experimentation cycle. This frequency modulated heat flux deposited over the sample generates thermal waves as a result of thermal oscillations at surface. A thermal wave of angular frequency ω has a wavelength λ, λ π (2α/ω) propagates with a velocity αω, in a medium of thermal diffusivity α. The amplitude of such a wave damps exponentially with depth, and its depth of penetration directly related to its wavelength λ in the given material, depends on its frequency, i.e., the lower the frequency, the larger is the wavelength and the deeper it propagates but with lower velocity. The depth at which energy of the wave attenuates to times of its surface value is called thermal diffusion length, which plays a vital role in thermography. Dependence of thermal diffusion length on frequency and band width of the applied signal facilitates the depth resolution of probing of different depths simultaneously, within a single sweep using FMTWI. Pulse compression with non-stationary signals prevalent in RADAR (RAdio Detection And Ranging) is designed to enhance target detection sensitivity and resolution. It concentrates the energy of the entire signal to about a short time instant, thereby producing a pseudo pulsed response from this long duration excitation and also enhances signal to noise ratio (SNR). This concept is extended in the field of Thermographic NDT considering the fact that the resultant thermal response on the surface at the location of defects (where the defects exist beneath the surface) differs not only in magnitude but also more importantly in delay depending on the depth of the defects [8-10]. In this work, a Hilbert transform based pulse compression approach considering the phase analysis along with cross correlation coefficient and time delay information between the thermal responses of different pixels with that of a chosen reference pixel profile is used to detect the subsurface anomalies [11]. This approach enables deeper probing of sub-surface defects with depth resolution.

3 Experimentation Experiments are carried out to validate the proposed method for detection d off voids on a mild steel sample containing six flat bottom holes of dimensions as shown in Fig. 1. Thee temporal temperaturet e history of the sample surface is captured byy the IR imaging system with a resolution off 320 x 256 pixels. The sample is exposed to heating by two 1 kw halogen lamps. Thee temperature responsee over the sample surface is captured at a frame rate of 20 Hz for duration of f 100 s and the sample response during the cooling process is usedd for the analysis. A linear frequency modulated signal (chirp) of 100s duration, with its frequency varying from 0.01 to 0.10 Hz, is generated from a signal generator and used to drive the heat sources via a source control unit, as shown in Fig. 2. a b c d e f Figure 1: Top and cross-sectional views of the experimental mild steel sample (all unitss are in cm) with six flat holes of same diameter (1cm) att different depths. Figure 2: Schematic illustration of experimental apparatus for FMTWI

8 sec respectivelyr are shown in Fig. 5. It can be seen s that the defects are visible inn both the images.

It is also observed that the visibility of the defects is inversely proportional to the depth of defect from the surface, i.e. lesser the depth of defect from surface more is the visibility.

defects. It was also observed that the defects show a time delay in comparison to the non-defective reference pixel.

filtering, the calculated amplitude and phase signals are used for defect detection.

The mean removed thermal response is used as input to the Hilbert based matched filter approach.

4 (a) (b) Figure 4: Illustration of the temporal temperature profile at a chosen non-defectivee referencee location. a) ). Transient temporal thermal profile. b) Mean removed thermal profile The phase and cross-correlation amplitude images at time instants 4.8 sec andd 20.8 sec respectivelyr are shown in Fig. 5. It can be seen s that the defects are visible inn both the images. However, the phase images provide slightly finer details of the defects in comparison to the amplitude images. It is also observed that the visibility of the defects is inversely proportional to the depth of defect from the surface, i.e. lesser the depth of defect from surface more is the visibility. Broadening due to lateral diffusion influence from the t deeper defects is also clear from the defects, more spreading s is observed with deeper defects. It was also observed that the defects show a time delay in comparison to the non-defective reference pixel. Results and Discussions The applicability of FMTWI for subsurface defect detection using the Hilbert transform based matched filter approach is demonstrated [ 12]. This approach mixes m the input temporal infrared signal with the in-phase and quadrature phase referencee Infrared signals and after low-pass filtering, the calculated amplitude and phase signals are used for defect detection. Thee block diagram of the proposed Hilbert approach is shown in Fig. 3. Figure 3: Block diagram of thee Hilbert transform based matchedd filtering Firstly, the temporal mean increase of each pixel thermal profile iss removed with the help of a first- order polynomial fitting, as shown in Fig. 4. The mean removed thermal response is used as input to the Hilbert based matched filter approach. The temporal infrared signal,, s(t) corresponding to each pixel is convolved with the chosen non-defective reference pixel, r(t) and its quadrature obtained by applying the Hilbert transform on the reference pixel. The cross-correlation fromm the in-phase of the reference signal. The phase of the cross-correlation signal is computed using g both the in-phase and quadrature phase parts. The reconstructed phase and cross-correlation images are filtered using average filter to remove any high frequency noise and it is observed that the smoothened images provide better visibility of defects. images are obtained from the amplitude part of the convolved signal obtained

5 (a) (b) (c) Figure 5:Infrared image capture by the IR camera, Phase and Cross-correlation amplitude images (of the cropped region) reonstructed from pulse compressed data of FMTWI, Pink circle annotations are used to highlightt the defects a) Infrared image captured c at 4.8 sec, with the pink rectangle depicting the region cropped for displaying the results, b)phase image (in degrees) at 4.8 sec, c) Cross correlation amplitude image at 20.8 sec Conclusion Defect detection in the mild steel sample having simulated artificial voidss of same diameter at different depths has been carried out by thee Frequency modulatedd thermal wave imaging using the Hilbert transform based matched filter method. Furthermore, it was w observed experimentally thatt each of the phase and correlation (height of cross-correlation peak p (amplitude) and location of cross-correlation peak (time delay)) images are capable of detecting sub-surface defects. Acknowledgement This work is partially funded by the t CSIR-CSIO network project, I-HEAL. I

6 References [1] Maldague, Xavier. "Applications of infrared thermography in nondestructive evaluation." Trends in optical nondestructive testing (2000): [2] Maldague, Xavier. "Theory and practice of infrared technology for nondestructive testing." (2001). [3] Balageas, Daniel L., Jean-Claude Krapez, and P. Cielo. "Pulsed photothermal modeling of layered materials." Journal of Applied Physics 59.2 (1986): [4] Spicer, Jane W. Maclachlan, et al. "Time-resolved infrared radiometry of multilayer organic coatings using surface and subsurface heating." Orlando'91, Orlando, FL. International Society for Optics and Photonics, [5] Busse, G., and P. Eyerer. "Thermal wave remote and nondestructive inspection of polymers." Applied Physics Letters 43.4 (1983): [6] Maldague, Xavier, and Sergio Marinetti. "Pulse phase infrared thermography." Journal of Applied Physics 79.5 (1996): [7] Mulaveesala, Ravibabu; Tuli, Suneet, "Theory of frequency modulated thermal wave imaging for nondestructive subsurface defect detection," Applied Physics Letters, vol.89, no.19, pp , , Nov 2006 [8] Ghali V S, Mulaveesala R and Takei M, Cross-correlation based compression technique for frequency modulated thermal wave imaging, QIRT Proceedings(2010), Quebec, Canada (2010),p-129. [9] Ghali, V. S., and R. Mulaveesala. "Frequency modulated thermal wave imaging techniques for non-destructive testing." Insight-Non-Destructive Testing and Condition Monitoring 52.9 (2010): [10] Mulaveesala, Ravibabu, Jyani Somayajulu Vaddi, and Pushpraj Singh. "Pulse compression approach to infrared nondestructive characterization." Review of Scientific Instruments 79.9 (2008): [11] Tabatabaei, Nima. Development of Frequency and Phase Modulated Thermal-wave Methodologies for Materials Non-destructive Evaluation and Thermophotonic Imaging of Turbid Media. Diss [12] Dua, G.; Mulaveesala, R., "Applications of barker coded infrared imaging method for characterisation of glass fibre reinforced plastic materials," Electronics Letters, vol.49, no.17, pp.1071, 1073, August

Sub-surface Thermal Wave Imaging

Sub-surface Thermal Wave Imaging Professor Suneet Tuli January 20, 2011 Contents 1 What is Thermal Non-Destructive Testing (TNDT) 1 2 Basic concepts of thermal imaging system 2 2.1 Stefan s law of radiation...............................