As part of the industry-wide effort to find suitable gasket materials for use in place of currently used asbestos containing gaskets, an experimental investigation was under- taken following an extensive literature search and review of the data from gasket material suppliers. This paper discusses the work to evaluate various asbestos/free gasket materials for steam turbines. The investigation was conducted in two phases. Phase I. Screening tests, consisted of mechanical property evaluations pf non-asbestos gasket materials after high temperature conditioning in mineral oil, water and high pressure steam environments. Based on the results of the Phase I tests, four materials were selected as potential replacements for asbestos containing gaskets. Phase II evaluated these four materials by more extensive testing under conditions closer to actual service conditions.

In view of the recent developments in finite element stress analysis and materials processing and testing, the reliability of fracture mechanics analysis on service life prediction is critically dependent on the estimation of the flaws by nondestructive test (NDT) methods. This paper discusses the correlation between NDT indications and actual sizes of natural flaws determined by destructively examining flaws and shows the importance of this uncertainty on component life predictions. A section of a low-alloy steel forging was ultrasonically inspected, saw-cut into cubes containing flaws, and examined by C-scan and X-ray. Successive metallographic polishing was performed to reconstruct the shape, size, and location of the flaws. The flaws were identified by using electron microprobe energy spectrum analysis. The flaw sizes predicted by ultrasonic inspection were underestimated by up to 6.6 times the actual flaw area. It was demonstrated that underestimation of the sizes of natural flaws by NDT should be properly corrected to make realistic cyclic life predictions.

The results of a systematic program undertaken to evaluate potential substitute materials for asbestoe-free gaskets used in power plant turbines are reviewed. Thirty-three commercially available materials were subjected to a series of tests reflecting the requirements of a good gasket material for turbine applications. The results identified flexible graphite as the only material to meet or exceed the properties of asbestos gaskets. One limitation of flexible graphite uncovered in the program, however, is its relatively high stress relaxation behavior especially at temperatures above 650°F (343°C). Ten-thousand-hour stress relaxation tests were conducted in the temperature range of 500 to 850°F (260 to 454°C) to determine the usable temperature limit of flexible graphite. The effects of joint flexibility and reloading were also evaluated at high temperatures.

The threshold stress intensity factor range ?Kth of three steels was studied in air and steam. Constant load and constant crack opening displacement (COD) tests were conducted by using tension-tension sinusoidal wave function loading. The COD tests were controlled by a high frequency clip-on strain gage. Crack growth rate was measured by both optical and COD compliance techniques and verified by metallographic examination of the broken specimens. The ?Kth value was determined as the asymptote to the best-fit data below a crack growth rate of 2.5 x 10-8 cm/cycle. The ?Kth values were higher in steam than in air. Crack growth rates in air and steam generally crossover or merge in the range of 2.5 x 10-6 to 2.5 x 10-7 cm/cycle; below this range the growth rate is lower in steam than in air. The higher values in steam tests were explained by environmentally assisted crack branching and corrosion products formed at the crack tip after long-time testing at growth rates of less than one lattice spacing per cycle of loading.