PSI - Issue 65

Maslov S.V. et al. / Procedia Structural Integrity 65 (2024) 139–146 Author name / Structural Integrity Procedia 00 (2024) 000–000

142

4

 



 

(4)

  ,  

Cr

Cr e

1     



e

where e Cr and α are the experimentally determined creep limit and creep rate of the strain gauge. Unfortunately, expression (4) can be directly used when processing the findings of SSS studies for full-scale structures only in rare cases, since the strain based on a strain gauge in the measurement process is a variable. Previously, when evaluating the creep of strain gauges during full-scale strain gauge studies, calculated values were taken for constant deformation corresponding to the maximum possible value for the considered time interval of structure loading. It is irrational to use such a method for studying the stress state of power plants under development, since at elevated temperatures and deformations, overestimation of the creep of strain gauges can lead to an unacceptably high estimate of the measurement error. Therefore, this paper proposes a combined method of estimation   Cr e    according to formula (2), in which this value is defined as a combination of possible values of uninformative components accumulated at separate time intervals of the test process. For this purpose, in addition to determining creep under constant deformation, it is proposed to use the results obtained under the action of deformation cyclically varying according to the sinusoidal law. For this purpose, a test bench is used in which a variable deformation is created in a cantilever beam heated by a flowing electric current, as shown in Fig. 1. a b

Fig. 1. Test bench for studying the creep of strain gauges: (a) – location of strain gauges; (b) – heating and loading of the specimen.

The strain gauges are located on two sides of the cantilever beam whose free end takes the vertical movements created by the rotating camshaft, as shown in Fig. 1a. The specimen is heated by passing an electric current of an adjustable magnitude through it (Fig. 1b). This results in alternating deformations based on strain gauges, varying in time according to a sinusoidal law. As shown by the results of experiments at different temperatures, amplitudes and frequencies of specimen vibrations, as the loading frequency rises, the amplitude of the strain gauge signals increases, as shown in Fig. 2a. This effect associated with the creep of strain gauges makes it possible to determine the creep limits for different test temperatures. For this purpose, graphs of creep dependences vs. inverse loading frequency are used (Fig. 2b). The asymptotes of the graphs correspond to the creep limits for temperatures of 520 °C and 650 °C and the maximum deformation amplitude in the cycle of 0.15%. Determining the creep rate α based on cyclic tests is a more difficult task (Vlasov D.D. et al, 2022) than in the case of using data obtained with constant deformation of specimens with strain gauges. To simplify the task, one can assume that this value is independent of the deformation level at a constant temperature. The analysis of experimental data showed that with a loading duration of up to 20–30 seconds, this assumption had practically no effect on the result obtained. To determine α, the solution of the inverse problem can be used by employing experimental data obtained at the same temperature for different levels of deformation. The function   ,    can be approximated according to a mesh scheme, and for the full use of the obtained data, it is advisable to use the least squares method.