- Email: [email protected]
Infrared radiation emitted due to scanning of a hot spot as a probe of hidden defects
Infrared Physics & Technology 76 (2016) 574–579
Contents lists available at ScienceDirect
Infrared Physics & Technology journal homepage: www.elsevier.com/locate/infrared
Infrared radiation emitted due to scanning of a hot spot as a probe of hidden defects Mariusz Woz´ny ⇑, Kinga Mas´, Serhiy Prokhorenko, Dariusz Ploch, E.M. Sheregii Centre for Microelectronics and Nanotechnology, University of Rzeszow, Pigonia 1, 35-959 Rzeszow, Poland
h i g h l i g h t s A scanning hot air nozzle is applied to introduce energy in a researched sample. The hidden defect has an increased temperature in comparison with the surround area. The scanning a controlled sample.
a r t i c l e
i n f o
Article history: Received 24 March 2016 Revised 4 April 2016 Accepted 8 April 2016 Available online 19 April 2016 Keywords: Thermography Defect Scanning Noninvasive control
a b s t r a c t Specially created subsurface defects in a sample are detected using a high resolution infrared camera FLIR SC7000. A scanning hot air (about 110 °C) nozzle is applied to introduce additional energy in a researched sample. The hidden defect has an increased temperature in comparison with the surrounding area that is a result of changed emissivity and thermal diffusivity. The suggested method is compared with pulse thermography which uses a xenon lamp for excitation. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction The non-contact method of temperature measurement by detecting radiation emitted from the surface in the entire spectral range is well-known as infrared (IR) thermal imaging or IR thermography (IRTG) [1]. Good progress in research on IR thermography to detect defects in the surface layer of materials has been observed in the last years. Thermography offers noncontact, wide area detection of material defects, and can be used as an alternative of or a complement to the conventional inspection technologies [2–9]. The essence of these researches is the thermal response analysis of a material stimulated by an external heat impulse. The presence of areas containing defects with thermal properties different from those in defect-free areas changes the diffusion rate that enables to observe the location of defects by analyzing the temperature distribution on the researched sample surface [2,7]. On the other hand, the size of defects detected using the pulse ⇑ Corresponding author. E-mail address: [email protected] (M. Woz´ny). http://dx.doi.org/10.1016/j.infrared.2016.04.005 1350-4495/Ó 2016 Elsevier B.V. All rights reserved.
thermography method should be not less than a few millimeters if a whole surface (usually the macroscopic one) of the researched sample is stimulated [8,9]. It seems that this paucity of method can be removed using a scanning laser source to deposit heat into a sample surface [3–5]. However, the IR thermography signal will be depended strongly from the optical characteristic of the surface in this case to a greater extent than on the subsurface defects. Other methods of excitation can be used in thermography to detect defects, namely ultrasound [10,11]. The ultrasound thermography uses the interaction between mechanical and thermal waves to detect material defects. If a damage in a component absorbs excited high energy ultrasound waves then it will locally heat up. The resulting temperature gradient is captured by an infrared camera on the sample surface. This method is suitable for many applications such as the crack detection, the delamination and impact damage control or adhesion and welded joints testing. In another words, it is also possible to detect some macroscopic discontinuities in materials using the ultrasound method [10,11]. This paper describes a new probe of subsurface defects of submillimeter sizes. It provides scanning over the controlled sample
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
surface with a specially designed hot air nozzle used as an excitation source, and then infrared radiation emitted from the sample is detected.
2. Theory The radiation emission from a tested body is the basis of thermography. According to the Stefan–Boltzmann law and taking into account the reflected ambient radiation as background and the self-radiation of the infrared thermometer, the electric signal of an IR detector is to be considered as follows [12]:
h i U ¼ C eT nsam þ ðe 1ÞT nbgr þ T ntrm
ð1Þ
where U is a detector signal, Tsam is the temperature of a measured sample; Tbgr is the temperature of background radiation; Ttrm is the temperature of an IR detector (IR camera); C is a constant specific for the IR detector; the exponent n depends on the wavelength k because the infrared thermometers do not cover the entire wavelength range: at wavelengths ranging from 1 to 14 lm n is between 17 and 2 (at long wavelengths it is between 2 and 3, and at short wavelengths it is between 15 and 17) [10]; r ¼ 1 e is reflection of the object, e = E/Es is emissivity of the sample surface material equal to the ratio of the emission intensity E of a real body to the emissivity Es of the absolute black body at the same temperature. Eq. (1) shows the basic correlation for non-contact temperature measurements. Furthermore, strong dependence of the detector signal U on the sample temperature T nsam (n P 4) allows registering
575
a change of the temperature distribution caused by some inhomogeneity (discontinuity) of the material properties. When the excitation energy is introduced the heat diffusion through measured material becomes important. Heat diffusion through a solid can be described by the Fourier’s law of the heat equation [13]. If a Dirac heat pulse is this excitation energy, a 1D solution of the Fourier equation can be found as the propagation of an ideal waveform defined as an intense unit-area pulse in a semi-infinite isotropic solid. Such solution has the form [14,15]:
Q TðtÞ ¼ T 0 þ pffiffiffiffiffiffi e pt
ð2Þ
where Q is the energy absorbed by the surface, and T0 is the initial temperature, e is the effusivity:
e¼
qffiffiffiffiffiffiffiffiffiffi kqcp
ð3Þ
which is a thermal property that determines the ability of a material to exchange heat with its surroundings and is an important parameter in this method. In Eq. (3) k is the thermal conductivity of the sample material, q is the material density, and cp is the heat capacity at constant pressure. The effusivity e is changed if the material properties (k, q, cp) are changed due to inhomogeneity (discontinuities of the material properties: inclusions, voids, delaminations and so on) that causes the local change of temperature DT and consequently, leads to the change of the detector signal DU. The time of this change is associated with another parameter – the thermal diffusivity a. Any
Fig. 1. (a) Scheme of the experiment; a hot air nozzle: hot air is getting out from the nozzle of 0.4 mm in diameter (blue color) and sucks up the air reflected from the investigated surface (white color); (b) Nozzle scheme.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
576
3. Materials and methods 3.1. Methods
It is the second and very important parameter of the proposed method which implies introducing the excitation energy in the form of a scanning hot spot, a reminiscent of the Dirac heat pulse. The relation between these two material parameters (e and a) enables to optimize the parameters of the method – optimal scanning velocity v for the hot spot as well as the minimal size of a subsurface defect which depends on the minimum DT value can be detected by the IR camera.
In order to obtain the Dirac heat pulse a scanning hot air nozzle is mounted on a stationary platform. The hot air flow is directed to the surface of the sample which is placed on the table moving with a predetermined speed in the direction of z axis (see Fig. 1a and b). The air pressure getting out of the nozzle was 0.2 MPa at the diameter of the outgoing air nozzle of 0.2 mm, the temperature of hot air was 110 °C. To ensure precise motion sequences of the sample, a motorized positioning system such as an OWIS PKTM 70 table movable in the x, y, z directions, was used. The positioning system provides choosing the sample velocity v with very high precision. A high resolution thermal imaging camera FLIR SC7000 (IR camera) is used to measure the temperature on the sample surface. High precision in determining the emission coefficient is essential for correct temperature measurements. This is particularly important if the researched surface is metallic because the emissivity is small in comparison with the reflectivity in this case. The IR camera used in our experiment operates in the short wave region of 3–5 lm. To perform correct measurements of the temperature distribution on the surface a series of the IR camera settings was made to verify the distance to a measured sample surface: a considerable portion of radiation from the object is absorbed in air. On the other hand, this distance influences the spatial resolution of the IR camera which is expected to be maximal at the distance of 0.3 m.
Fig. 3. (a)The U(z)-curves obtained by scanning along the yellow line (in Fig. 2) using the radiation from the regions on the heat trace shown below (b) using the radiation from the regions out (but close) of the heat track, shown above.
Fig. 4. The U(z)-curve obtained by scanning along the blue line (see Fig. 2a) using the radiation from the region of the heat trace, shown in Fig. 3b. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 2. Picture of the sample: (a) front size, (b) back size, and (c) scale.
change of the sample temperature Tsam is related to the thermal diffusivity a [13],
a ¼ k=qc
ð4Þ
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
3.2. Sample The sample of 60 mm 20 mm 5 mm in size was made of C45 steel. Five holes of 6 mm, 5 mm, 4 mm, 3 mm, 2 mm in diameter were purposely made on the back side of the sample to imitate defects. The depth of holes is equal to two thirds of the sample thickness. Front and back sides of the sample are shown in Fig. 2a and b. The scale is shown in Fig. 2c in order to illustrate the positions of holes on the back side. 3.3. Optimization of parameters As mentioned above, the velocity v should be optimized taking into account its relation with the heat diffusivity a. The heat pffiffiffiffiffiffi diffusion length LH ¼ as is determined by the thermal diffusivity a, and the next inequality should be fulfilled:
dP
pffiffiffiffiffiffi
as
577
ð5Þ
where d is the defect size and s is the time of the nozzle movement along the defect area s = d/v. The nonequation (5) means that the time s is less than the heat diffusion time at the same distance. Therefore, the nozzle velocity is limited by minimum
vP
a d
ð6Þ
On the other hand, this velocity also has a top limit because the defect must be registered using the change of temperature
Q DT ¼ pffiffiffiffiffiffi e pt
ð7Þ
as it was shown by Eq. (2). The top limit for the nozzle velocity depends on the minimum value Q necessary for the efficient temperature change in the defect area; simultaneously the effusivity also
Fig. 5. Quasi-3D-distribution of detected signal U(y, z): defects shown in Fig. 2 (back side of the sample) are seen as maxima obtained through scanning the front side along different traces by the heat spot – in two courses are shown.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
578
influences the DT increment. Thus, the nozzle velocity maximum depends on the material parameters of the examined sample and on the Q value delivered by the nozzle. The last one can be derived from the nozzle velocity v, the hot air temperature Tair and the hot air flux velocity Vair = DV/Dt (DV is the volume of the hot air getting out from the nozzle during the time interval Dt), namely:
Q ¼ cair qair ðT air T 0 Þ
V air d
v
ð8Þ
where cair and qair are the hot air heat capacity and the hot air density, respectively. So, the top limit for the nozzle velocity can be derived from the relations (7) and (8) using the DTmin value which is possible to detect by the IR camera as the DU(y, z). The best nozzle velocity v selected during the measurements was equal to 4.8 mm/s (also it is the speed of the OWIS PKTM 70 table). 3.4. Measurement procedure The measurement consisted of registering the signal U(z) as a function of the coordinate value z through scanning the sample surface by the hot air nozzle along the z axis at the fixed y value. The FLIR SC7000 camera registered the radiation generated by the heat trace during scanning with the hot spot. It is important to choose a proper region of the sample surface to register radiation by IR camera. This region should be within the heat trace on the surface. The choice is facilitated by Altair, the IR camera software. The signal intensity associated with different regions of a sample can be displayed by this computer program. Thus, the registered magnitude U(z) depends on the intensity of radiation generated by the area selected on the heat trace. 4. Results Figs. 3 and 4 present the experimental curves of U(z) registered under the above described experiment conditions. Fig. 3a presents the U(z)-curves obtained during scanning along the red line where a row of the holes is placed on the back side of the sample (see Fig. 2a–c). The heat trace regions selected to register the radiation are shown below the curves (Fig. 3b). In Fig. 3a it can be seen that the U(z)-curve displays the maxima corresponding to increased temperature at the positions of defects (holes shown in Fig. 2b). Whereas, when the U(z)-curve is obtained by measuring the IRsignal from the region out of the heat trace (see Fig. 3b), the maximum intensity at the positions of the largest holes is much lower, and no maxima are observed at the positions of the last two holes. It means that the detection of hidden defects depends on which region is selected as a source of the IR-signal, namely: this region should be within the heat trace made by the nozzle. Fig. 4 shows the U(z)-curve obtained during scanning along the green line (see Fig. 2a) which is out of the holes row. No maxima are displayed on this curve, though the region to register the radiation was selected on the heat trace. This confirms the method effectiveness of. Furthermore, scanning along different traces on the front side of the sample allows building quasi-3D-images of registered defects as a distribution of the function U(y, z). It is shown in Fig. 5. Fig. 6 shows the temperature distribution T(y, z) for scanning the sample surface by the heat spot. It is seen that the temperature increases at the hole positions on the back side. It is clear that the DTmin = 0.03 °C.
Fig. 6. The temperature distribution on the sample surface while scanning by the heat spot.
(subsurface) defects in material because even a specially made defect shaped as a hole of 2 mm in diameter (the hole deep equal to 2/3 of the sample thickness) on the back side of the sample is detected by scanning the front side of the sample with the aid of a hot spot. For comparison, an impulse of a xenon flash lamp was chosen as an alternative excitation source. In this experiment the impulse of 6 kJ energy and the 6 ms duration generated the power enough to register the temperature distribution on the surface of the sample and to visualize hidden defects by the IR camera according to Wysocka-Fotek et al. [15]. As shown in Fig 7, after the excitation of the sample by the xenon flash lamp the IR-camera registered only the first three largest holes with diameters d P 3 mm on the back side of the sample. The last two smallest holes are invisible when this method is used. Whereas, scanning by the hot air nozzle allowed detecting all five holes, with the minimal size being 2 mm. Thus, the proposed method of hot spot scanning is more effective for the detection of defects smaller than 4 mm. The method enables to deliver the energy directly to a defect area that makes the material more sensitive to heat exchange with its surrounding as determined by the thermal effusivity e (see Eqs. (2), (3)).
5. Discussion The above experimental results show evidently that the described experimental procedure is sensitive to hidden
Fig. 7. (a) Image of the sample back side in the visible region of spectra, (b) thermogram of the same surface obtained by the IR-camera after impulse excitation using a xenon lamp: three holes are visible clearly, and (c) scale.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
Eq. (2) points out that the highest temperature is in the region where more energy Q is delivered. If the nozzle moves along the larger defect, according to Eq. (8), the Q value is getting larger too, because it is proportional to d – size of the defect. It seems worthy to analyze a minimum size of a defect which can be detected by our method. A hidden hole (made on the back side of another sample) with a diameter of 0.5 mm was detected experimentally. Theoretically, this minimum size can be calculated taking into account maximum sensitivity of an IR-camera to the temperature increment DT while registering the radiation intensity from a sample surface. This maximum sensitivity is 0.01 K according to the IR-camera specifications confirmed experimentally as was shown above. It is expected, that a hole with a diameter of 0.2 mm may be registered using the proposed method under described experimental conditions. 6. Conclusion The infrared radiation generated by the scanning heat spot can be used to detect hidden (subsurface) defects. The specimen of C45 steel with flat-bottom holes was prepared. The proposed method of noncontact and noninvasive control of hidden defects using the scanning hot air nozzle, enables to detect the holes located on the back side of the steel plate (which imitate a hidden defect) of d 6 2 mm in diameter. The alternative noncontact and noninvasive method using a xenon flash lamp impulse as a source of excitation enables to detect the holes on the same steel plate with a diameter of d P 3.5 mm. Theoretically, a minimum size of a hidden defect which allows defect detecting with the proposed method, is 0.2 mm. This assumption will be verified in further experiments. Thus, the impulse thermography able to detect macroscopic defects can be followed by the thermography using a scanning hot spot to check some ‘‘suspicious” places where defects below 2 mm in size are expected. We believe that the thermography using a flash lamp and scanning the suspected places by a hot air nozzle could compete with a standard 2-step procedure of defect control applied in today’s industry (X-radiation defectoscope and fluorescent penetration).
579
Acknowledgments Financial support of Structural Funds in the Operational Programme – Innovative Economy (IE OP) financed from the European Regional Development Fund – Project ‘‘Modern material technologies in aerospace industry”, Nr POIG.01.01.02-00-015/08-00 is gratefully acknowledged. References [1] M. Vollmer, K.-P. Möllman, Infrared Thermal Imaging, Wiley-Vch, NY, 2010. [2] X. Maldague, Theory and Practice of Infrared Technology for Nondestructive Testing, John Wiley – Interscience, NY, 2001. [3] S.E. Burrows, S. Dixon, S.G. Pickering, T. Li, D.P. Almond, Thermographic detection of surface breaking defects using a scanning laser source, NDT&E Int. 44 (2011) 589. [4] S.A. Morris, Detection and characterization of package defects and integrity failure using dynamic scanning infrared thermography (DSIRT), J. Food Sci. 81 (2016) E388. [5] Teng Li, Darryl P. Almond, D. Andrew S. Rees, Crack imaging by scanning pulsed laser spot thermography, NDT&E Int. 44 (2011) 216. [6] X. Maldague, S. Marinetti, Pulse phase infrared thermography, J. Appl. Phys. 79 (1996) 2694. [7] Yuxia Duan, Stefanie Huebner, Ulf Hassler, Ahmad Osman, Clemente IbarraCastanedo, Xavier P.V. Maldague, Pulse phase thermography for defect detection and visualization, Infrared Phys. Technol. 60 (2013) 275. [8] F.C. Sham, C. Nelson, L. Long, Surface crack detection by flash thermography on concrete surface, Insight 50 (2008) 240. [9] Masashi Ishikawa, Hiroshi Hatta, Yoshio Habuka, Ryo Fukui, Shin Utsunomiya, Detecting deeper defects using pulse phase thermography, Infrared Phys. Technol. 57 (2013) 42. [10] Th. Zweschper, A. Dillenz, G. Riegert, G. Busse, Ultrasound thermography in NDE: principle and applications, Acoustical Imaging 27 (2004) 113. [11] A. Gleiter, Ch. Spießberger, G. Busse, Lockin thermography with optical or ultrasound excitation, J. Mech. Eng. 56 (2010) 619. [12] Basic Principles of Non-Contact Temperature Measurement. Innovative Infrared Technology, Optris, Berlin, 2015. [13] H.S. Carslaw, J.C. Jaeger, Conduction of Heat in Solids, Clarendon Press, Oxford, 1959. [14] M. Pilla, M. Klein, X. Maldague A. Salerno, New absolute contrast for pulsed thermography, in: D. Balageas G. Busse, C. Carlomagno (Eds.), Proc. QIRT (Quantitative Infrared Thermography), Dubrovnik, Croatia, 2002. [15] O. Wysocka-Fotek, W. Oliferuk, M. Maj, Reconstruction of size and depth of simulated defects in austenitic steel plate using pulsed infrared thermography, Infrared Phys. Technol. 55 (2012) 363.
Contents lists available at ScienceDirect
Infrared Physics & Technology journal homepage: www.elsevier.com/locate/infrared
Infrared radiation emitted due to scanning of a hot spot as a probe of hidden defects Mariusz Woz´ny ⇑, Kinga Mas´, Serhiy Prokhorenko, Dariusz Ploch, E.M. Sheregii Centre for Microelectronics and Nanotechnology, University of Rzeszow, Pigonia 1, 35-959 Rzeszow, Poland
h i g h l i g h t s A scanning hot air nozzle is applied to introduce energy in a researched sample. The hidden defect has an increased temperature in comparison with the surround area. The scanning a controlled sample.
a r t i c l e
i n f o
Article history: Received 24 March 2016 Revised 4 April 2016 Accepted 8 April 2016 Available online 19 April 2016 Keywords: Thermography Defect Scanning Noninvasive control
a b s t r a c t Specially created subsurface defects in a sample are detected using a high resolution infrared camera FLIR SC7000. A scanning hot air (about 110 °C) nozzle is applied to introduce additional energy in a researched sample. The hidden defect has an increased temperature in comparison with the surrounding area that is a result of changed emissivity and thermal diffusivity. The suggested method is compared with pulse thermography which uses a xenon lamp for excitation. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction The non-contact method of temperature measurement by detecting radiation emitted from the surface in the entire spectral range is well-known as infrared (IR) thermal imaging or IR thermography (IRTG) [1]. Good progress in research on IR thermography to detect defects in the surface layer of materials has been observed in the last years. Thermography offers noncontact, wide area detection of material defects, and can be used as an alternative of or a complement to the conventional inspection technologies [2–9]. The essence of these researches is the thermal response analysis of a material stimulated by an external heat impulse. The presence of areas containing defects with thermal properties different from those in defect-free areas changes the diffusion rate that enables to observe the location of defects by analyzing the temperature distribution on the researched sample surface [2,7]. On the other hand, the size of defects detected using the pulse ⇑ Corresponding author. E-mail address: [email protected] (M. Woz´ny). http://dx.doi.org/10.1016/j.infrared.2016.04.005 1350-4495/Ó 2016 Elsevier B.V. All rights reserved.
thermography method should be not less than a few millimeters if a whole surface (usually the macroscopic one) of the researched sample is stimulated [8,9]. It seems that this paucity of method can be removed using a scanning laser source to deposit heat into a sample surface [3–5]. However, the IR thermography signal will be depended strongly from the optical characteristic of the surface in this case to a greater extent than on the subsurface defects. Other methods of excitation can be used in thermography to detect defects, namely ultrasound [10,11]. The ultrasound thermography uses the interaction between mechanical and thermal waves to detect material defects. If a damage in a component absorbs excited high energy ultrasound waves then it will locally heat up. The resulting temperature gradient is captured by an infrared camera on the sample surface. This method is suitable for many applications such as the crack detection, the delamination and impact damage control or adhesion and welded joints testing. In another words, it is also possible to detect some macroscopic discontinuities in materials using the ultrasound method [10,11]. This paper describes a new probe of subsurface defects of submillimeter sizes. It provides scanning over the controlled sample
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
surface with a specially designed hot air nozzle used as an excitation source, and then infrared radiation emitted from the sample is detected.
2. Theory The radiation emission from a tested body is the basis of thermography. According to the Stefan–Boltzmann law and taking into account the reflected ambient radiation as background and the self-radiation of the infrared thermometer, the electric signal of an IR detector is to be considered as follows [12]:
h i U ¼ C eT nsam þ ðe 1ÞT nbgr þ T ntrm
ð1Þ
where U is a detector signal, Tsam is the temperature of a measured sample; Tbgr is the temperature of background radiation; Ttrm is the temperature of an IR detector (IR camera); C is a constant specific for the IR detector; the exponent n depends on the wavelength k because the infrared thermometers do not cover the entire wavelength range: at wavelengths ranging from 1 to 14 lm n is between 17 and 2 (at long wavelengths it is between 2 and 3, and at short wavelengths it is between 15 and 17) [10]; r ¼ 1 e is reflection of the object, e = E/Es is emissivity of the sample surface material equal to the ratio of the emission intensity E of a real body to the emissivity Es of the absolute black body at the same temperature. Eq. (1) shows the basic correlation for non-contact temperature measurements. Furthermore, strong dependence of the detector signal U on the sample temperature T nsam (n P 4) allows registering
575
a change of the temperature distribution caused by some inhomogeneity (discontinuity) of the material properties. When the excitation energy is introduced the heat diffusion through measured material becomes important. Heat diffusion through a solid can be described by the Fourier’s law of the heat equation [13]. If a Dirac heat pulse is this excitation energy, a 1D solution of the Fourier equation can be found as the propagation of an ideal waveform defined as an intense unit-area pulse in a semi-infinite isotropic solid. Such solution has the form [14,15]:
Q TðtÞ ¼ T 0 þ pffiffiffiffiffiffi e pt
ð2Þ
where Q is the energy absorbed by the surface, and T0 is the initial temperature, e is the effusivity:
e¼
qffiffiffiffiffiffiffiffiffiffi kqcp
ð3Þ
which is a thermal property that determines the ability of a material to exchange heat with its surroundings and is an important parameter in this method. In Eq. (3) k is the thermal conductivity of the sample material, q is the material density, and cp is the heat capacity at constant pressure. The effusivity e is changed if the material properties (k, q, cp) are changed due to inhomogeneity (discontinuities of the material properties: inclusions, voids, delaminations and so on) that causes the local change of temperature DT and consequently, leads to the change of the detector signal DU. The time of this change is associated with another parameter – the thermal diffusivity a. Any
Fig. 1. (a) Scheme of the experiment; a hot air nozzle: hot air is getting out from the nozzle of 0.4 mm in diameter (blue color) and sucks up the air reflected from the investigated surface (white color); (b) Nozzle scheme.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
576
3. Materials and methods 3.1. Methods
It is the second and very important parameter of the proposed method which implies introducing the excitation energy in the form of a scanning hot spot, a reminiscent of the Dirac heat pulse. The relation between these two material parameters (e and a) enables to optimize the parameters of the method – optimal scanning velocity v for the hot spot as well as the minimal size of a subsurface defect which depends on the minimum DT value can be detected by the IR camera.
In order to obtain the Dirac heat pulse a scanning hot air nozzle is mounted on a stationary platform. The hot air flow is directed to the surface of the sample which is placed on the table moving with a predetermined speed in the direction of z axis (see Fig. 1a and b). The air pressure getting out of the nozzle was 0.2 MPa at the diameter of the outgoing air nozzle of 0.2 mm, the temperature of hot air was 110 °C. To ensure precise motion sequences of the sample, a motorized positioning system such as an OWIS PKTM 70 table movable in the x, y, z directions, was used. The positioning system provides choosing the sample velocity v with very high precision. A high resolution thermal imaging camera FLIR SC7000 (IR camera) is used to measure the temperature on the sample surface. High precision in determining the emission coefficient is essential for correct temperature measurements. This is particularly important if the researched surface is metallic because the emissivity is small in comparison with the reflectivity in this case. The IR camera used in our experiment operates in the short wave region of 3–5 lm. To perform correct measurements of the temperature distribution on the surface a series of the IR camera settings was made to verify the distance to a measured sample surface: a considerable portion of radiation from the object is absorbed in air. On the other hand, this distance influences the spatial resolution of the IR camera which is expected to be maximal at the distance of 0.3 m.
Fig. 3. (a)The U(z)-curves obtained by scanning along the yellow line (in Fig. 2) using the radiation from the regions on the heat trace shown below (b) using the radiation from the regions out (but close) of the heat track, shown above.
Fig. 4. The U(z)-curve obtained by scanning along the blue line (see Fig. 2a) using the radiation from the region of the heat trace, shown in Fig. 3b. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 2. Picture of the sample: (a) front size, (b) back size, and (c) scale.
change of the sample temperature Tsam is related to the thermal diffusivity a [13],
a ¼ k=qc
ð4Þ
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
3.2. Sample The sample of 60 mm 20 mm 5 mm in size was made of C45 steel. Five holes of 6 mm, 5 mm, 4 mm, 3 mm, 2 mm in diameter were purposely made on the back side of the sample to imitate defects. The depth of holes is equal to two thirds of the sample thickness. Front and back sides of the sample are shown in Fig. 2a and b. The scale is shown in Fig. 2c in order to illustrate the positions of holes on the back side. 3.3. Optimization of parameters As mentioned above, the velocity v should be optimized taking into account its relation with the heat diffusivity a. The heat pffiffiffiffiffiffi diffusion length LH ¼ as is determined by the thermal diffusivity a, and the next inequality should be fulfilled:
dP
pffiffiffiffiffiffi
as
577
ð5Þ
where d is the defect size and s is the time of the nozzle movement along the defect area s = d/v. The nonequation (5) means that the time s is less than the heat diffusion time at the same distance. Therefore, the nozzle velocity is limited by minimum
vP
a d
ð6Þ
On the other hand, this velocity also has a top limit because the defect must be registered using the change of temperature
Q DT ¼ pffiffiffiffiffiffi e pt
ð7Þ
as it was shown by Eq. (2). The top limit for the nozzle velocity depends on the minimum value Q necessary for the efficient temperature change in the defect area; simultaneously the effusivity also
Fig. 5. Quasi-3D-distribution of detected signal U(y, z): defects shown in Fig. 2 (back side of the sample) are seen as maxima obtained through scanning the front side along different traces by the heat spot – in two courses are shown.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
578
influences the DT increment. Thus, the nozzle velocity maximum depends on the material parameters of the examined sample and on the Q value delivered by the nozzle. The last one can be derived from the nozzle velocity v, the hot air temperature Tair and the hot air flux velocity Vair = DV/Dt (DV is the volume of the hot air getting out from the nozzle during the time interval Dt), namely:
Q ¼ cair qair ðT air T 0 Þ
V air d
v
ð8Þ
where cair and qair are the hot air heat capacity and the hot air density, respectively. So, the top limit for the nozzle velocity can be derived from the relations (7) and (8) using the DTmin value which is possible to detect by the IR camera as the DU(y, z). The best nozzle velocity v selected during the measurements was equal to 4.8 mm/s (also it is the speed of the OWIS PKTM 70 table). 3.4. Measurement procedure The measurement consisted of registering the signal U(z) as a function of the coordinate value z through scanning the sample surface by the hot air nozzle along the z axis at the fixed y value. The FLIR SC7000 camera registered the radiation generated by the heat trace during scanning with the hot spot. It is important to choose a proper region of the sample surface to register radiation by IR camera. This region should be within the heat trace on the surface. The choice is facilitated by Altair, the IR camera software. The signal intensity associated with different regions of a sample can be displayed by this computer program. Thus, the registered magnitude U(z) depends on the intensity of radiation generated by the area selected on the heat trace. 4. Results Figs. 3 and 4 present the experimental curves of U(z) registered under the above described experiment conditions. Fig. 3a presents the U(z)-curves obtained during scanning along the red line where a row of the holes is placed on the back side of the sample (see Fig. 2a–c). The heat trace regions selected to register the radiation are shown below the curves (Fig. 3b). In Fig. 3a it can be seen that the U(z)-curve displays the maxima corresponding to increased temperature at the positions of defects (holes shown in Fig. 2b). Whereas, when the U(z)-curve is obtained by measuring the IRsignal from the region out of the heat trace (see Fig. 3b), the maximum intensity at the positions of the largest holes is much lower, and no maxima are observed at the positions of the last two holes. It means that the detection of hidden defects depends on which region is selected as a source of the IR-signal, namely: this region should be within the heat trace made by the nozzle. Fig. 4 shows the U(z)-curve obtained during scanning along the green line (see Fig. 2a) which is out of the holes row. No maxima are displayed on this curve, though the region to register the radiation was selected on the heat trace. This confirms the method effectiveness of. Furthermore, scanning along different traces on the front side of the sample allows building quasi-3D-images of registered defects as a distribution of the function U(y, z). It is shown in Fig. 5. Fig. 6 shows the temperature distribution T(y, z) for scanning the sample surface by the heat spot. It is seen that the temperature increases at the hole positions on the back side. It is clear that the DTmin = 0.03 °C.
Fig. 6. The temperature distribution on the sample surface while scanning by the heat spot.
(subsurface) defects in material because even a specially made defect shaped as a hole of 2 mm in diameter (the hole deep equal to 2/3 of the sample thickness) on the back side of the sample is detected by scanning the front side of the sample with the aid of a hot spot. For comparison, an impulse of a xenon flash lamp was chosen as an alternative excitation source. In this experiment the impulse of 6 kJ energy and the 6 ms duration generated the power enough to register the temperature distribution on the surface of the sample and to visualize hidden defects by the IR camera according to Wysocka-Fotek et al. [15]. As shown in Fig 7, after the excitation of the sample by the xenon flash lamp the IR-camera registered only the first three largest holes with diameters d P 3 mm on the back side of the sample. The last two smallest holes are invisible when this method is used. Whereas, scanning by the hot air nozzle allowed detecting all five holes, with the minimal size being 2 mm. Thus, the proposed method of hot spot scanning is more effective for the detection of defects smaller than 4 mm. The method enables to deliver the energy directly to a defect area that makes the material more sensitive to heat exchange with its surrounding as determined by the thermal effusivity e (see Eqs. (2), (3)).
5. Discussion The above experimental results show evidently that the described experimental procedure is sensitive to hidden
Fig. 7. (a) Image of the sample back side in the visible region of spectra, (b) thermogram of the same surface obtained by the IR-camera after impulse excitation using a xenon lamp: three holes are visible clearly, and (c) scale.
M. Woz´ny et al. / Infrared Physics & Technology 76 (2016) 574–579
Eq. (2) points out that the highest temperature is in the region where more energy Q is delivered. If the nozzle moves along the larger defect, according to Eq. (8), the Q value is getting larger too, because it is proportional to d – size of the defect. It seems worthy to analyze a minimum size of a defect which can be detected by our method. A hidden hole (made on the back side of another sample) with a diameter of 0.5 mm was detected experimentally. Theoretically, this minimum size can be calculated taking into account maximum sensitivity of an IR-camera to the temperature increment DT while registering the radiation intensity from a sample surface. This maximum sensitivity is 0.01 K according to the IR-camera specifications confirmed experimentally as was shown above. It is expected, that a hole with a diameter of 0.2 mm may be registered using the proposed method under described experimental conditions. 6. Conclusion The infrared radiation generated by the scanning heat spot can be used to detect hidden (subsurface) defects. The specimen of C45 steel with flat-bottom holes was prepared. The proposed method of noncontact and noninvasive control of hidden defects using the scanning hot air nozzle, enables to detect the holes located on the back side of the steel plate (which imitate a hidden defect) of d 6 2 mm in diameter. The alternative noncontact and noninvasive method using a xenon flash lamp impulse as a source of excitation enables to detect the holes on the same steel plate with a diameter of d P 3.5 mm. Theoretically, a minimum size of a hidden defect which allows defect detecting with the proposed method, is 0.2 mm. This assumption will be verified in further experiments. Thus, the impulse thermography able to detect macroscopic defects can be followed by the thermography using a scanning hot spot to check some ‘‘suspicious” places where defects below 2 mm in size are expected. We believe that the thermography using a flash lamp and scanning the suspected places by a hot air nozzle could compete with a standard 2-step procedure of defect control applied in today’s industry (X-radiation defectoscope and fluorescent penetration).
579
Acknowledgments Financial support of Structural Funds in the Operational Programme – Innovative Economy (IE OP) financed from the European Regional Development Fund – Project ‘‘Modern material technologies in aerospace industry”, Nr POIG.01.01.02-00-015/08-00 is gratefully acknowledged. References [1] M. Vollmer, K.-P. Möllman, Infrared Thermal Imaging, Wiley-Vch, NY, 2010. [2] X. Maldague, Theory and Practice of Infrared Technology for Nondestructive Testing, John Wiley – Interscience, NY, 2001. [3] S.E. Burrows, S. Dixon, S.G. Pickering, T. Li, D.P. Almond, Thermographic detection of surface breaking defects using a scanning laser source, NDT&E Int. 44 (2011) 589. [4] S.A. Morris, Detection and characterization of package defects and integrity failure using dynamic scanning infrared thermography (DSIRT), J. Food Sci. 81 (2016) E388. [5] Teng Li, Darryl P. Almond, D. Andrew S. Rees, Crack imaging by scanning pulsed laser spot thermography, NDT&E Int. 44 (2011) 216. [6] X. Maldague, S. Marinetti, Pulse phase infrared thermography, J. Appl. Phys. 79 (1996) 2694. [7] Yuxia Duan, Stefanie Huebner, Ulf Hassler, Ahmad Osman, Clemente IbarraCastanedo, Xavier P.V. Maldague, Pulse phase thermography for defect detection and visualization, Infrared Phys. Technol. 60 (2013) 275. [8] F.C. Sham, C. Nelson, L. Long, Surface crack detection by flash thermography on concrete surface, Insight 50 (2008) 240. [9] Masashi Ishikawa, Hiroshi Hatta, Yoshio Habuka, Ryo Fukui, Shin Utsunomiya, Detecting deeper defects using pulse phase thermography, Infrared Phys. Technol. 57 (2013) 42. [10] Th. Zweschper, A. Dillenz, G. Riegert, G. Busse, Ultrasound thermography in NDE: principle and applications, Acoustical Imaging 27 (2004) 113. [11] A. Gleiter, Ch. Spießberger, G. Busse, Lockin thermography with optical or ultrasound excitation, J. Mech. Eng. 56 (2010) 619. [12] Basic Principles of Non-Contact Temperature Measurement. Innovative Infrared Technology, Optris, Berlin, 2015. [13] H.S. Carslaw, J.C. Jaeger, Conduction of Heat in Solids, Clarendon Press, Oxford, 1959. [14] M. Pilla, M. Klein, X. Maldague A. Salerno, New absolute contrast for pulsed thermography, in: D. Balageas G. Busse, C. Carlomagno (Eds.), Proc. QIRT (Quantitative Infrared Thermography), Dubrovnik, Croatia, 2002. [15] O. Wysocka-Fotek, W. Oliferuk, M. Maj, Reconstruction of size and depth of simulated defects in austenitic steel plate using pulsed infrared thermography, Infrared Phys. Technol. 55 (2012) 363.