CBRN event in Switzerland on Tue 20th August 1974

On August 20, 1974 at 11.20 a.m. a trip on turbine TG-1 occurred resulting in high bearing and casing vibrations (bearing 6:60). At trip time, generator 2 was delivering about 140 MVar. Resulting from a failure of the steam dump system to operate, with the consequence that the relief valve did not open. That resulted in a rapid rise of coolant temperture, steam pressure and pressurizer level and pressure. At 160 bar of pressure in the primary, the pressurizer pressure relief valves opened, lowering rapidly the pressure in the primary. About 10 seconds after valve opening, the pressure had reached such a low level that the pressurizer pressure relief valves were reactuated to close. Due to a disturbance, valve PCV-456, failed to close, resulting in a lowering of RCS pressure up to 100 bar after about 1 minute. Reactor tripped resulting from a low pressure signal (126.5) bar. Due to the opening of the pressurizer relief valve, the pressure in RCS dropped to about 70 bar, corresponding to a saturation temperature of 284 deg. C. Consequently, steam appeared in the primary hot leg, filling the pressurizer. Two or 3 minutes after trip, the operator recognized that failure of the relief valve and isolated it with the power operated valve 531. The water level began to drop, and 11 minutes after trip, automatic SI was initiated by low pressure and level in the pressurizer. Si systems worked normally and about 40 litres per second of water were spilled through the four SI pump nozzles into the primary, causing a rise of pressure to 110 bars and a further rise of level to 70%. The SI pumps were then turned of and the power operated valves of the spray pipings were closed. From that moment on, the pressurizer level could be controlled through charging pumps and release of steam, assuming the primary to cool down. About 3 minutes after trip, the containment pressure alarm signal was actuated because of too high pressure, and 1 minute later the high activity alarm. Maximum pressure in containment reached 100 mbar over normal. The operators activated the containment fan coolers. Since several safety alarms of the pressurizer relief tank were on, it was quickly assumed that the rupture disc was broken that the discharge channel was defectuous. After TG-1 trip, due to steam dump failure, steam pressure rose to 66 bar. After TG-2 trip, following reactor trip, steam pressure rose to over 70 bar, acutating the safety valves and thus lowering pressure to about 65 bar.
Analysis of the Causes of the Incident
The trip is not unfamiliar and would not have affected the primary if steam dump had normally been actuated. An inspection of containment after primary had cooled down, showed that the yoke between the PCV-456 valve housing and air engine was broken, and probably due to a dynamic effort on the piping at opening of the valve. When reviewing the sequence of events, the failure of two systems, namely the steam dump and the pressurizer relief system, we came to the conclusion that it did not bring to an uncontrollable nor a damaging situation. During the incident, no activity (in gas or liquid form) in the surrounding area reached an uncontrollable level. The generator safety valves maintained the steam pressure within allowable limits. The SIS brought back the primary to a safer pressure, allowing normal cooldown conditions. Water out of primary system = 1,8 m3
Proposal for Modifications: Steam Dump System
a) Revisions and calibrations should be made in steam dump system (before opening of steam dump valve.) b) Studies will be made, to make periodic controls of steam dump while in operation. It should help to insure better safety limits (for example: unwanted opening of steam dump valve.) c) A control type writer lined to the steam dump will be installed in order to control the opening of steam dump valves and to check the good working of oil pumps.
BASIS FOR RATING: Initiator: possible-small LOCA, real. Safety function availability: full.
A-2R = 1/2, 1 chosen because the operator closed the leak timely.
Final level: 1+1 = 2. Point 4.2 in users' manual page 21 d) procedures to control steam dump system function has been considered inadequate, upgrading +1.
DIFFICULTIES IN RATING: If the initiator is expected (reactor coolant system leakage) the rating also would be 0/1 = level 1 or 2