PSI - Issue 5

Yui Izumi / Structural Integrity Procedia 00 (2017) 000 – 000

4

Yui Izumi et al. / Procedia Structural Integrity 5 (2017) 683–688

686

PC

Infrared camera

Cold insulator

Low temp.

High temp.

Halogen lump

Specimen

Fatigue crack

Aluminium

Aluminium

Heat insulating material

Fig. 4. Schematic illustration of the experimental setup of temperature gap measurement.

2.2. Temperature gap measurement

A schematic illustration of the experimental setup of temperature gap measurement was shown in Fig. 4. Temperature gradient in the specimen longitudinal direction was set to 0.04  C/mm based on the field temperature condition of steel bridge. For thermal image measurement, the high performance infrared camera (FLIR systems, phoenix infrared camera) with InSb infrared sensors (temperature resolution: 25 mK, spectral range: 3.0 - 5.0  m, spatial resolution: 320  256 pixels) was employed. Flat black paint for emissivity improvement was applied on the temperature measurement area. Fig. 5 shows the temperature evaluation area. Results of temperature gap measurement in the specimen No.1 show Figs. 6(a), (c) and (b), (d) are respectively the temperature distribution and the distribution in the x -direction temperature differential, and Fig. 6 (c) and (d) show the results of noise reduction processing by spatial smoothing (3  to the raw temperature data. In the distribution of raw temperature differential as shown in Fig. 6(b), there is no clear temperature gap at the crack. However, for the distribution of smoothed temperature differential as shown in Fig. 6(d), it can be seen that temperature gap appears around the back surface fatigue crack indicating thermal insulation effect by the crack. Fig. 7 is the temperature differential distributions of each specimen obtained by spatial smoothing (9  . In the specimen No.1, No.2 and No.4, it is found in the temperature differential distribution that temperature gap clearly appears at the crack. For the specimen No.3, there is no clear temperature gap at the crack, because crack width is small. 3. Experimental results

58 mm

60 mm

Fig. 5. Temperature evaluation area.