Abstract
The service life of high-pressure hydrogen storage vessels at fueling stations is dictated by fatigue crack growth. Standards such as the ASME Boiler and Pressure Vessel Code (BPVC) provide a methodology for calculating the fatigue-limited design life of high-pressure hydrogen storage vessels, in which one essential input is the fatigue crack growth rate (da/dN) vs. stress-intensity factor range (ΔK) relationship measured for the material of construction in the service environment, i.e. hydrogen gas. These measurements must also be conducted at sufficiently slow cyclic loading frequency since decreasing the frequency usually results in faster crack growth rates. Generation of complete fatigue crack growth data sets according to standard test methods becomes time consuming and expensive when these environment and frequency criteria are met. Two modifications to standard test procedures are shown to reduce the time and associated costs related to this testing. One approach is to accelerate the rate at which ΔK changes with respect to crack extension. This can be accomplished by controlling the normalized K-gradient, C, where C= 1/K•dK/da. In addition, by using negative values of C, (i.e. decreasing ΔK) multiple da/dN vs ΔK segments can be generated from a single test specimen. This paper presents results from an experimental effort to identify the limits to which these two strategies may be employed to measure fatigue crack growth relationships for pressure vessel steels in gaseous hydrogen environments. Critical to this effort is understanding and preventing load history effects than can result in non-conservative fatigue crack growth measurements. Specifically, the use of a variable, rather than constant, value of C is shown to be necessary to maximize the efficiency of fatigue crack growth measurement and avoid load history effects under decreasing K. Steeper K-gradients, and more efficient testing, can be achieved by measuring crack closure and subsequently adjusting ΔK to the closure-corrected effective value, ΔKeff, using the adjusted compliance ratio (ACR) method. This project aims to apply these results to develop procedures for efficiently measuring full fatigue crack growth relationships, including the near threshold region, over a broad range of load ratio.