Papers by Author: Woo Gon Kim

Abstract: This study aimed to assess creep crack growth rates on the base and welded metals of modified 9Cr-1Mo steel. For this purpose, welded specimens were prepared by the Shielded Metal Arc Weld method. To obtain mechanical properties concerning the base and welded metals, a series of creep and tensile tests were conducted at 600 °C, and creep crack growth tests were also performed under different applied loads using 1/2" compact tension specimens at 600 °C. Their creep crack growth rates were calculated using the empirical equation of the da/dt vs. C* parameter and compared. It appeared that, for a given value of C*, the rate of creep crack propagation was about 2.0 times faster in the welded metal than the base metal. This reason was that the welded metal was faster in the creep strain rate than the base metal.

Abstract: In this paper, a series of statistical studies were conducted on creep crack growth behavior of Grade 9Cr-1Mo steel for next generation reactor. Creep crack growth tests were performed on pre-cracked compact tension (CT) specimens under the applied load ranges from 3800 to 5000N at the identical temperature condition of 600oC. The creep crack growth behavior has been analyzed statistically using the empirical equation between crack growth rate da/dt and C* parameter, namely da/dt=B(C*)q. First, the determination methods of B and q obtained from experiments were investigated by the least square fitting method and the mean value method. The probability distribution functions of B and q have been investigated using the normal, log-normal and Weibull distribution. The constant B and q are followed well 2-parameter Weibull. Second, the creep crack growth rate data were generated by Monte-Carlo simulation method assuming the 2-parameter Weibull in B and q parameters. The probability distribution of creep crack growth rate for arbitrary C* parameter values seems to follow well Weibull distribution.

Abstract: Creep behavior for Alloy 617, which is considered as one of the major structural materials of a very high temperature reactor, was investigated in air at 950oC. Creep experimental data was obtained by a series of creep tests with different stress levels at 950oC. Alloy 617 revealed little primary creep strains and unclear secondary creep stages. A tertiary creep stage was initiated from a low strain level and was dominant in full creep curves. The creep constants of A, n, m, and C in Norton’s power law and Monkman-Grant relationships were determined. In microstructure observations of crept specimens, it was found that a Cr2O3 oxidation layer was formed on the surface, and just beneath the Cr2O3 layer, an internal Al-oxide sub layer was formed with rod shapes. Also, below the internal sub layer, a thick carbide-depleted zone was developed due to reaction of the chromia and carbide precipitates. The thickness of the outer Cr-oxide layer increased with increasing creep rupture times. The increasing tendency showed a smooth slope like a parabolic curve.

Abstract: An isochronous stress-strain curve (ISSC) needs to be generated for a creep design application for high-temperature materials. To generate the ISSC for type 316LN stainless steel (SS), a series of creep data, which was obtained from creep tests with different stress levels at 600oC, was used. Creep curves were modeled by means of a nonlinear least square fitting (NLSF) of the Garofalo model. In the fitting of the creep curve, a secondary creep region was separated into first and second phases, and its fitting range was suitable to use for the first phase. The Garofalo model revealed a good agreement with the experimental creep data, and its parameters, P1, P2 and P3 revealed a good linear relationship as a function of a stress. The ISSCs for type 316LN SS at 600oC were successfully generated up to 300,000 hours.

Abstract: This study aimed to model the long-term creep curves above 105 hours by implementing a nonlinear least square fitting (NLSF) of the Kachanov-Rabotnov (K-R) model. For this purpose, the short-term creep curves obtained from a series of creep tests at 950oC were used. In the NLSF of their full creep curves, the K-R model represented a poor match to the experimental curves, but the modified K-R one revealed a good agreement to them. The Monkman-Grant (M-G) strain represented the behavior of a stress dependency, but the 􀁏 parameter was constant with a stress independency. The 􀁏 value in the modified K-R model was 2.78. Long-term creep curves above 105 hours from short-term creep data were modeled by the modified K-R model.

Abstract: To design HTGR components for up to 1000oC, their creep curves are necessary during a design process. In this study, the full creep curves were modeled by the nonlinear least square fitting method using the Kachanov-Rabotnov (K-R) creep model. A series of creep data was obtained experimentally under various stress levels for Hastelloy-X at 950oC, and the data was used to model the creep curves. The K-R model gave a poor description of modeling creep curves, but the modified K-R one, which has another variable, K in the K-R model, was in better agreement than the K-R one. It was found that the λ parameter in the K-R model was constant regardless of the stress variations. The λ value was about 3.9 for the K-R model and about 5.8 for the modified one.

Abstract: The creep properties of type 316LN SS welded by the SAW process have been evaluated. The creep tests for both the base and weld metals were conducted with different stress levels at 550oC and 600oC. The creep-rupture time of the weld metal did not show a large difference when compared to that of the base one, though it exhibited a little lower value at 600oC. The creep rate of the weld metal was lower than that of the base one at the same stress and rupture-time conditions. Especially, the creep-rupture ductility of the weld metal is found to be decreased by about 60%, compared to the base one. This is due to the decreasing of tensile elongation and the increasing of the yield stress in the weld metals.

Abstract: The creep properties for the Hastelloy-X alloy which is one of candidate alloys for a high temperature gas-cooled reactor are presented. The creep data was obtained with different stresses at 950oC, and a number of the creep data was collected through literature surveys. All of the creep data were combined together to obtain the creep constants and to predict a long-term creep life. In the Norton’s creep law and the Monkman-Grant relationship, the creep constants, A, n, m, and m’ were obtained. Creep master curves based on the Larson-Miller parameter were presented for the standard deviations of 1σ, 2σ and 3σ. Creep life at each temperature was predicted for a longer-time rupture above 105 hours. Failure probability was also estimated by a statistical process of all the creep rupture data.

Abstract: This paper presents the results of the Minimum Commitment Method (MCM) applied to predict the creep rupture life of type 316LN SS. Constant A, and the function of P(T) and G(σ) being used in the MCM equation were determined. To determine a proper value of the constant A, a focal point method and a trial and error one were adopted, respectively. It was found to be A=-0.02~-0.05 for type 316LN SS. Each prediction curve with the A values were presented up to 106 hours and compared to the experimental data at each temperature. Using the short-term creep rupture data for under 2,000 hours, a long-time rupture reaching up to 106 hours was predicted by the MCM.

Abstract: To predict the creep-rupture life of type 316LN stainless steels which are major structural components of liquid metal reactors, a number of creep-rupture data were collected through literature survey and experimental data of KAERI. Using the data, the creep-rupture life was analyzed by means of the Larson-Miller, the Orr-Sherby-Dorn and the Manson-Haferd parametric methods. Polynomial equations for predicting the creep life were obtained. In order to analyze the acceptance and use of the parametric methods, standard error values were accurately investigated by statistical process of the creep data. As for the results, the three parametric methods are found to be favorable in predicting the creep life of type 316LN stainless steel. Each method did not generate a large error in the standard error of the estimate with variations of the temperatures, but the Orr-Sherby-Dorn and the Manson-Haferd methods showed a better agreement than the Larson-Miller one. Especially, at higher the 700oC, the Manson-Haferd method conformed well to the experimental data. The reason is because the Manson-Haferd method includes two constants of ta and Ta.