Abstract
The following conclusions have been derived from the present analytical and experimental studies on the creep of a cylindrical tube of low carbon steel subjected to internal pressure at 450°C. In the cylindrical tube with an arbitrary radius ratio K, subjected to combined axial load and internal pressure, it was found that the following assumption was efficiently suitable for the creep analysis of the tube in the transient or steady state creep stage. The present assumption is the constancy of both the equivalent stress σav and the stress ratio A at the mean radial distance of the cylindrical tube during the creep, where the stress ratio A was defined as the ratio of average tangential stress σθav to the axial stress σzav.
The first of the present verification was made through the aspect of the influences of K and A on the analytical results of the creep of the cylindrical tube. It was concluded that the analytical results were in good agreement with the experimental results. The second of the present verification was performed from both the types of the creep tests of the tubular specimens under the periodic stress variation along the single Mises-stress ellipse and under that along the two Mises-stress ellipses. It was also concluded that the analytical results were closely like the results of the experiments made under the general loading condition of both the variation of axial load and of internal pressure during the creep tests. Therefore, it was found that both the influences of the change in the equivalent stress at the mean radial distance of the tube during the creep and of the multiaxiality of the stress or the stress ratio at the mean radial distance were not predominant in the creep analysis of the high pressure tube. By such simple assumption, the creep strain distributions along the radial distance of the tube were easily calculated from the information of simple tensile creep. It should be noted, however, that the present analytical result gives a slightly higher estimation of creep strain than the measured values in the case of variation of the equivalent stress.