Thermal Dimensional Stability of Low Thermal Expansion Ductile Cast Irons

• Main Authors: Zong-Pei Liu, 劉宗霈
• Other Authors: 潘永寧
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/01144556435646396570

碩士 === 國立臺灣大學 === 機械工程學研究所 === 102 === The objectives of this study are threefold: (1) To investigate the effect of homogenization heat treatment on both the degree of Ni segregation and the content of dissolved carbon in the matrix of the low thermal expansion ductile cast irons, and then to analyze the influence of compositional factor on α value; (2) To study the effects of alloy composition and homogenization heat treatment on the dimensional and shape changes of the test specimens by means of constrained thermal cyclic fatigue tests (30~200℃); (3) The temperature distribution and the thermal stress field in the test specimens after the constrained thermal cyclic fatigue tests were analyzed first by both calculation and simulation (ANSYS), and then the dimensional and shape changes of the alloys studied (Alloy D-5, Regular ductile iron and 304 stainless steel) were calculated and compared with the measured data. In addition, the correlations among α value, thermal stress and dimensional change were evaluated. The experimental results indicate that the degree of Ni segregation can be reduced by increasing the homogenization heat treatment temperature and/or time, rendering a decrease in α value. On the other hand, the dissolved C content in the matrix showed little affected by homogenization heat treatment, regardless of the fact that increasing both heat treatment temperature and time will increase the dissolved C content. Among the various heat treatment procedures employed, heat treatment T4(1150℃/4hr/FC/750℃/4hr) not only can effectively eliminate the Ni segregation, but also can reduce the C concentration in the matrix, resulting in a very low α value of around (2-3)×10-6/℃. Furthermore, regression equations were derived to correlate the degree of Ni segregation and dissolved C content with α value, as expressed below: Heat A (35%Ni): Heat B (30%Ni + 5%Co): Heats A &; B: It is clear from the above equations that in order to achieve a low α value, both the degree of Ni segregation and the dissolved C content should decrease. Constrained thermal cyclic fatigue tests (30~200℃) were performed to compare the dimensional and shape changes among the low thermal expansion ductile cast irons with different homogenization heat treatment procedures, and also among three different alloys studied herein, namely, Alloy D-5, Regular ductile iron and 304 stainless steel. The extent of distortion or shape change of the test specimens (&;#8710;PV) was used as a criterion to evaluate the dimensional stability of the alloys investigated. The effect of homogenization heat treatment on &;#8710;PV can be expressed by the following order: Heat A (35%Ni): T0(338.51μm)>T1(301.82μm)>T2(237.95μm)>T3(190.70μm)>T4(61.24μm) Heat B (30%Ni + 5%Co): T0(286.24μm)>T1(252.64μm)>T2(189.44μm)>T3(125.46μm)>T4(48.23μm) It is clear from the above results that Alloy A with heat treatment T4 (1150℃/4hr/FC/750℃/4hr) exhibits the lowest shape change (48.23μm), implying that an alloy with a lower α value can achieve a better dimensional stability. Finally, numerical simulation by finite element method (FEM) was employed to obtain the temperature distribution and thermal stress field for different alloys (D-5, regular SG and SUS304) after thermal cyclic fatigue tests. The results show that the values of temperature gradient follow the following order: SG>SUS304>D-5, while the order of the thermal stress is: SUS304>SG>D-5. Furthermore, regression analysis was performed to obtain the correlation between thermal stress and &;#8710;PV, with the results shown as follows: . It is obvious that the lower the thermal stress, the lower the &;#8710;PV value. In conclusion, low thermal expansion ductile iron with T4 heat treatment (1150℃/4hr/ FC/750℃/4hr/WQ) exhibits the best dimensional stability due to its lowest α value (1.72x10-6/℃).