PSI - Issue 13

Isabela Procopio et al. / Procedia Structural Integrity 13 (2018) 97–103 I. Procopio/ Structural Integrity Procedia 00 (2018) 000 – 000

101

5

The INCEFA+ fatigue tests are used to determine the effect of surface roughness, strain amplitude, mean stress/strain and hold time on the corrosion fatigue life of 304 stainless steel subjected to an applied strain range (or occasionally stress range) for a relatively low numbers of cycles (low cycle fatigue). 4.1. Hold time and mean stress/strain Specification of conditions for hold time and stress/strain was particularly demanding. Some considerations about these issues will be made next. Whereas conventional fatigue testing usually applies continuous triangular or sawtooth cyclic loading, this does not represent power plant operation (in which long periods at relatively constant load are alternated with occasional larger transients). The aim of the hold time tests is therefore to assess the extent to which the fatigue lives measured in standard cyclic fatigue tests are relevant to plant conditions. The proposed plan for hold time testing for INCFA-PLUS is to perform the tests at a temperature of 300 ⁰ C, with strain amplitudes of 0.3% and 0.6%, using a waveform of sawtooth (with faster load drop to help test efficiency). This way, the project can address conflicting data in open literature and perform tests that consider plant relevant holds in a manageable test duration. The planned hold frequency (with a maximum of 3 holds per test) is shown in Table 1:

Table 1. Planned hold frequency

Number of cycles Every 6000 cycles Every 1200 cycles Every 1000 cycles Every 200 cycles

Air/LWR

Strain amplitude (%)

Air

0.3 0.3 0.6 0.6

LWR

Air

LWR

The majority of INCEFA+ fatigue tests is performed under strain control with mean strain for Phase I testing. Strain amplitude values of 0.3% and 0.6% were planned, with either zero or 0.5% mean strain during cycling, Table 2.

Table 2. Mean strain

ε a (%)

Mean ε (%) for cycling

ε range (%) -0.6 to 0.6 -0.1 to 1.1 -0.3 to 0.3

Hold ε (%)

0.6 0.6 0.3 0.3

0

0

0.5

0.5

0

0

0.5

0.2 to 0.8

0.5

No effect of mean strain has been seen in the strain control tests (0.5% mean strain was applied). This result was anticipated and provides a useful comparison against already available data for similar tests in strain control. Exploration of mean stress effects under both strain and stress control is carried out in parallel to the main Phase 2 testing. 4.2. Phases of the project The project has three testing phases: phase 1 started in December 2016, phase 2 in May 2018 and Phase 3 is being defined. At the time of writing this paper, 56 (of 74) tests have been performed. The remaining tests are expected to be finished by August 2018. Sixty-five (65) of them are on a common 304L structural material relevant to current European nuclear power plants. The tests have been performed in accordance with a “Test Protocol” ( Vankeerberghen et al. (2018)), developed by the consortium, that determines the specimen size, the temperature, humidity level and pressures during the tests and the procedures to identify and store the specimens, among others. An expert panel has been created with the aim of reviewing all data generated by the project to ensure high data quality. So far, there is insufficient data to reveal any effect of hold time; so, further tests will be performed in Phase 2 to confirm this information. No effect of mean strain (0,5% mean strain has been applied) in strain control tests have