[en] The PHWR plant Atucha II had to be evaluated via a probabilistic risk assessment (PRA) by KWU. This part of the paper describes the objective, the performance and the results of the analysis as well as the evaluation of the safety concept of the 745 MWe PHWR of KWU. (orig.)

[en] Experiments designed to simulate certain types of postulated loss-of-coolant accidents in CANDU reactors show that, for some conditions, long time delays occur before the injected water is able to penetrate down the feeders to reach the horizontal fuel channels. The time delays are identified as a ''hot-wall'' effect, caused by countercurrent flow flooding in the feeders, due to the interaction between the water downflow and the upflow of steam generated by release of heat from the feeder pipe walls. This paper describes results of full-scale tests to measure the hot wall delays for the CANDU reactor core geometry. A theoretical model is described that predicts the measured time delays reasonably well. (orig.)

[en] The state-of-the-art experimental work in several countries on LOCA simulation burst and reflood tests in PWR rod bundles is reviewed concerning the influence of thermal-hydraulics on fuel clad ballooning and coolability of blocked fuel rod bundles. Representative two-phase flow heat transfer during reflooding limits the mean total circumferential burst strains to approx. 50%. Reversed coolant flow direction from the refilling to the reflooding phase which is typical of a combined emergency core cooling injection results in a maximum flow blockage of approx. 50%. With unidirectional flow the maximum blockage amounts up to approx. 70%. Reflood tests in partially blocked bundles have shown that coolability of PWR fuel rod bundles blocked up to 90% can be maintained without any unacceptable temperature increase due to the flow blockage. (orig.)

[en] A non-linear multi-node computer code used for steady-state modelling of vertical natural circulation U-tube steam generators, for PWR plants, is described. The methods employed for the physical and mathematical representation of steam generator performance, numerical solution techniques and empirical correlations for two phase flow and heat transfer are related. Aim of the paper is the determination of pressure, temperature and steam quality for the fluids in both primary and secondary sides. Particular attention is devoted to the analysis of the heat transfer coefficients in the boiling region. The method proposed is very simple and has the merit of predicting reliable results with very small computational time. (orig./HP)

[en] To avoid duplication, comments concentrate on the following topics: thermohydraulics and fuel behavior research for transients including LOCA; physical phenomena and system behavior during severe accidents; material issues of the pressure boundary and secondary circuit, probabilistic methods in reactor safety; optimization of instrumentation and control, including man-machine interaction; interaction between licensing and research. (orig./HP)

[en] This article presents an analysis of the transient behaviour. The analysis has been performed by means of simulation programs which are generated by the DSNP program system (''Dynamic Simulator for Nuclear Power Plants''). The simulation includes the reactor core, the heat removal system comprising several heat removal loops with different design and the turbine/generator set. Several perturbations of the heat removal balance caused by different types of initiating events are investigated. (orig./HP)

[en] The reference CANDU 600, designed under Canadian licensing criteria, has been evaluated for LOCA, under LWR acceptance criteria as applied in Japan. Offsite power is assumed unavailable following a LOCA and reactor trip. We show that the 1200⁰ peak cladding temperature criterion can be met with modest changes to the mechanical shutoff rods. The second shutdown system (poison injection), meets the temperature criterion without change. To speed core refill in the absence of forced circulation, the ECC system uses a directed injection to the unbroken end of the core. Other inherent CANDU safety features such as the availability of the moderator as a heat sink for LOCA without ECC injection, are retained. (orig.)

[en] Oxidation effects on CANDU fuel after a LOCA are significant only if the fuel cladding temperature exceeds about 1000⁰C. To reach such temperatures effectively requires the flow to stagnate in th reactor core. This requires a specific break size. The duration of low flow depends on the thermohydraulic characteristics of the circuit. The stagnation influences fuel clad temperatures, but to first order, the temperature is determined by the operating temperature, hence power, of the fuel. For the CANDU reactor second order effects on the cladding temperature arise from the power transient, metal water reaction, blowdown heat removal and radiation. The high temperature duration is short, i.e. less than a minute. We show that radiative heat transfer from the fuel to the pressure tube is an important inherent mechanism for limiting metal water reaction rates. Further, we establish that the duration and temperature of the transient is well within established embrittlement criteria. (orig./HP)

[en] In support of the Pressurized Thermal Shock Integration Study the Idaho National Engineering Laboratory has performed analyses of overcooling transients using the RELAP5/MOD1.6 and MOD2.0 computer codes. These analyses were performed for the H. B. Robinson Unit 2 pressurized water reactor, which is a Westinghouse 3-loop design plant. Results of the RELAP5 computer code as a tool for analyzing integral plant transients requiring a detailed plant model, including complex trip logic and major control systems, are examined. (orig.)

[en] The main feature of the fine motion control rod drive is the ball nut-spindle system inside the guiding tube enabling continuous insertion of the control rod into the reactor core. The mechanical drive of the spindle is an electrically powered motor transmitting the corresponding revolutions over a gear to the spindle. For reactor scram motion of hydraulic system is used. It comprises a high pressure nitrogen-water reservoir connected by water lines to the control rod housing. In case of demand a fast opening valve allows water to flow into the bottom of the housing to move a piston which in turn moves the control rod upward. Completely satisfactory results have been obtained with this system over 16 years. (orig.)