Steam generator tube failure event occurred at Ulchin unit 4 on April 5, 2002. Ulchin unit 4 was in shutdown mode to initiate cooling down at 01:20 on April 5, 2002 for the 3rd refueling outage. At 18:33 (all times are KST) the operators observed notable decreases in the pressurizer level and pressure at the beginning stage of cooling down. In reponse, the letdown flow was isolated and a third charging pump started to compensate the loss of primary coolant inventory at 18:55. Even though a third charging pump was activated, it couldn't make up the loss. While the operators suspected a steam generator tube leak, the radiation monitors on steam generator #2 blowdown lines alarmed at 18:46. At the same time, the operators isolated the affected steam generator using the Emergency Operating Procedure (EOP). At 18:49, the operators actuated high pressure safety injection (HPSI) system which was blocked just before the cooling down. The pressurizer level was recovered by the HPSI flow and the reactor was safely cooled down using the intact steam generator. At 19:59, the primary and secondary pressures were equalized. At 13:25 on the following day, the reactor reached a cold shutdown. The failed tube was visually examined just after the event. The failure was located on the hot leg side of the tube in Row 14, Column 38 at the top of tube sheet. The R14-C38 is near the center of the tube bundle. The failure consisted of a longitudinal split 75mm long in fish mouth opening shape from 3mm above the tubesheet to 78mm high and a circumferential rupture that completely severed the tube at the top of longitudinal split.For metallurgical inspection, the failed tube was removed from the B steam generator. The circumferentially severed section showed that there were two tearing areas. The one is located at the junction area with longitudinal split at the top of tube sheet and the other in the final ligament that seemed to rupture when the ligament could not sustain the transient loads. The severance propagated in helical way through the direction of 45 degree against the horizontal direction. This indicates that the circumferential severance developed after the longitudinal failure occurred as the secondary effect. The failure in longitudinal direction was in a fish-mouth type opening and the maximum opening was located in the middle of the ruptured tube. The longitudinal failure is thought to propagate in both ways, up and down. The upper propagation went through the circumferential severance changing its orientation about 45o to the circumference of the tube. The lower propagation arrived at the top of the tube sheet and continued in the circumferential direction to about 260o of the tube circumference. The failure shape is a “T” type that is a combination of normal fish-mouth opening in axial direction and circumferential severance.The rupture was caused mainly by SCC developed in the longitudinal direction from the top of tube sheet to the location of circumferential severance on the inside diameter of the tube. The cracks are shown to penetrate through wall of the tube in about the all along the longitudinal failed section. Several secondary cracks branched from the main axial crack were also observed. At the near longitudinal crack, a deformed area in a banded shape were observed along the crack faces. So far, the root cause is not clear, but this banded area seems to provide some primary effect on developing stress corrosion cracks.The metallography of the failed tube showed typically well structured grains of Inconel 600 HTMA. From this metallography, it was difficult to conclude that the material of failed tube is susceptible to stress corrosion cracking. In the past the SCC test of the same material demonstrated that the material was resistant enough to SCC. No radiation releases to the environment due to this event were found and there were no impacts on reactor safety.