PSI - Issue 37

Jan Kec et al. / Procedia Structural Integrity 37 (2022) 598–605 Jan Kec / Structural Integrity Procedia 00 (2019) 000 – 000

601

4

Fig. 3. Schematic illustration of the unique pipe testing system

The HBM QuantumX MX1615B data acquisition system was used for strain gauge measurements. The data from the pressure sensor, strain gauges and temperature sensors were measured at a rate of 1 Hz. HBM Type 1-LY11-6/120 strain gauges with a 6 mm grid size and self-compensating to steel were placed on tested pipes. The position of the strain gauges and the measured wall thickness is shown in Figs. 1 and 2. Strain gauges T1 to T4 were placed on the bare pipe to the position with the highest/lowest measured wall thickness. Strain gauges T5 to T8 were placed on the surface of the sleeves. 2.2. Results and discussion The pressure record during static loading can be seen for both pipes in Fig. 4. During the first bulge, the pressure increase was stopped and the last hold was performed. At this point, the pressure drop occurred due to low temperature creep above the yield strength attributed to sliding of existing dislocations and movement of new dislocations, see Wang et al., 2001 for a more detailed discussion of this field of interest. The pressure drop at maximum dwell for DN500 and DN700 pipe was 0,5 and 0,4 MPa respectively. The strains on the bare pipe can be seen in Fig. 5. For the DN500 pipe, an almost linear response to pressure was observed with a maximum value of 1600 µm/m measured on the circumferentially oriented strain gauge T1, which was placed at the lowest measured pipe wall thickness. The differences in strain between the lowest and highest wall thickness (T1T2 vs T3T4) are negligible, see Fig. 5a. The opposite was the case for the strain analysis on the DN700 bare pipe, where a significant difference between the lowest to the highest wall thickness can be observed, see Fig. 5b. On the circumferentially oriented strain gauge T1 high plastic deformation developed during the dwell. In general, the bulging of the two tubes was very inhomogeneous and is probably due to the inhomogeneity of the wall thickness. The strains on the CL sleeve can be seen in Fig. 6a. It can be seen that the response of both pairs of strain gauges is linear up to a pressure of about 9 MPa. The values of the maximum deformation are one third lower than on the bare pipe, suggesting good reinforcement by the sleeve. Unusual behaviour can be observed on the CS sleeve, see Fig. 6b. A pair of circumferential (T5 and T7) and longitudinal (T6 and T8) strain gauges show a considerable difference in behaviour, both in character and in absolute values. On the circumferential strain gauges (T7 and T9), a curvilinear response to load at lower pressures up to 4 MPa can be seen, the shape of the curve is convex for the T5 strain gauge while it is concave for the T7 strain gauge. Above 4 MPa, the strain dependence starts to linearize. When the maximum pressure is reached, a difference in the strain value obtained for both circumferential strain gauges is observed. The longitudinal strain gauge T6 had a linear response up to the maximum pressure, but the strain on the strain gauge T8 increased up to a pressure of 4 MPa along the curve and then started to linearize. The orientation of the strain also differed, with strain taking on positive values on strain gauge T6 and negative values on strain gauge T8. From the stress-strain curves, it can be estimated that the reinforcing effect of the CS sleeve becomes more effective at higher pressures.