PSI - Issue 14

Altaf Ali et al. / Procedia Structural Integrity 14 (2019) 273–281 Author name / Structural Integrity Procedia 00 (2018) 000–000

277

5

IQI region is 160 µm as third IQI wire is visible usually. So, in the weld region, the threshold for the smallest material thickness variation in the direction of X-ray path can be found using equation (1) and comes out to be 174.26 µm. Various parameters like unpenetrated height (L up ), length along circumference or circumferential coverage denoted by β, and its orientation with respect to X-rays (represented by α) can affect LOP detection in RT (as shown in Fig. 3 (c) below). These parameters affect the depth of material loss in travelling X-ray path and hence final grey value. Chances of detection also depend upon axial width of LOP along the pin axis. However, in this theoretical analysis, it has not been taken into account and is being assumed that LOP is in the plane of X-ray propagation. As an example, material loss depth variation with L p is shown in Fig. 3 (d) below for L up = 10% (% of tube wall thickness). As can be noticed that for lower value of α (α= 5 0 ) material loss in X-ray path crosses the threshold of 174.26 µm in weld region. On the other hand, same defect with same geometry ceases to cross the threshold at higher value of α (α= 30 0 ) and hence its chances detection becomes less. This shows that LOP detection in RT depends on its orientation with respect to X–rays. However, for this kind of defect geometry and weld configuration, UT can be used with quit good reliability. This study has been carried out to check its feasibility for the same.

Fig. 3. (a) Histogram showing SNR distribution in the weld zone; (b) Histogram showing SNR distribution in the IQI zone; (c) Schematic representations of the parameters affecting LOP detection in RT; (d) variation of ‘material loss depth’ encountered by X-rays for 10% LOP defect for two different orientations 4. UT experimental In order to have information about various dimensions of the weld zone after end plug TIG welding process, 52 numbers of setup and process metallography weld samples were analysed so that UT scanning time and distance can be optimized. Two important dimensions of the weld zone are shown in Fig. 1 (b) above. II tells about distance of clad end plug interface from starting point of weld zone and ‘I’ gives information of total length of weld zone to be scanned. Statistical distribution of dimension I shows that the total length (along the pin axis) of weld is centered about 2.6 mm and the maximum length observed is around 3.7 mm (See Fig. 4 (c) & (d)). So a scanning length of about 4 mm would be sufficient for full weld coverage. Critical angle analysis for steel water interface using snells’s law shows that first and second critical angle are 14.47 0 and 27.09 0 , Schmerr (2016). From reflection coefficient variation with incidence angle, an angle of incidence of 18.78 0 can be used that produces 450 shear wave in the material. For immersion UT, only shear wave mode has been used for better interpretation of results. Fig. 4 (b) shows schematic of refracted and incident waves in weld