PSI - Issue 64

André Weber et al. / Procedia Structural Integrity 64 (2024) 1959–1966 A. Weber/ Structural Integrity Procedia 00 (2019) 000 – 000

1965

7

Fig. 4: Advanced combined concept showing creep rupture strength and residual strength Thus, the representative load during the test would be a load that corresponds to the action increased by the safety factor for the action. However, a residual strength corresponding to at least 1.3 times the applied load would then have to be verified. For this purpose, samples that have already reached, for example, 50%, 80% or another percentage of their lifetime shall be subjected to a residual strength test. As this lifetime shall be representative for the whole service life of the material the load shall be reduced compared to the known failure line. Experience has shown that the residual strength in this case is always significantly higher than the stress during the creep test. In many cases it is similar to the strength in the undamaged state if the time to failure is not reached by a small margin. In logarithmic scale less then a decade could be chosen. According the scattering of the failure points this means a 10-20 % reduction of the load compared to the characteristic long term load for the wanted time (for example 3000 h). The bar is loaded under sustained load and environmental action for this time and a residual strength test has to be performed hereafter. The residual strength testing must now show that the material safety factor is satisfied. That means that the residual strength must be higher than the stress during ageing. For this residual strength testing there is no connection or reduction factor compared to the virgin strength needed. There is the comparison to the stress while ageing only. 3.5. Discussion There are different testing concepts. First of all if a certain level has to be shown to be safe, a simple long term test can be chosen. It seems logical that if higher sustained levels will be allowed in a guideline, that long term testing has to follow to higher levels. If the performance is not (yet) known a time to failure concept may be a better concept. A combination of both of them should be taken into account, if the performance is known very well and the performance shall be optimized. This is a testing regime close to the proposed ASTM or CSA testing regime. The