From 28 to 29 May 2004 a scheduled plant shutdown for repairing a primary leakage (gland of a control valve ) inside the containment was carried out. In the course of the restart on 28 May (criticality at 22 h 48), at about 50°C, the reactor was made critically for warming-up. During that process, which is closely monitored by the operator to switch over timely the neutron flux monitor ranges (eight decades) to prevent a Scram, he did not fully recognize the positive temperature coefficient of the fuel near to the end of the cycle (power increased to approx. 4% of rated power and a water temperature of 142 °C). This caused a significantly higher warm up rate than it is specified in the Technical Specification, which has to be controlled in periods of 30 min. Although the operator recognized the excessively high heat-up rate within that period and cooled down the reactor coolant system, the average rate of 56°C/h was violated. The following evaluation of consequences (stress calculation), required by the Technical Specification as a condition for further proceeding with the warm-up was not done timely. The calculation showed later that no negative impact of the event occurred.