Today there is a consensus that the current system thermal-hydraulic computer codes are not sufficiently validated for some conditions and for some phenomena relevant to natural circulation (low pressure, low driving heads, effect of non-condensables, boron transport at low velocities, etc. ). It is important to have assurance that the natural circulation systems are effective during all sequences in which they are required to function, and to define the needs for computer codes capable of taking into account all the important phenomena. Extensive work is needed to develop such codes and high quality experimental data are necessary to validate them. Therefore, separate effect and integral tests may be required both for additional validation of existing and/or development of new computer codes. However the targets for the computer codes accuracy should be clearly defined before asking for or designing new experiments to be sure that the financial resources required for the experiments are spent effectively.

In many cases, sufficient operating experience for the natural circulation systems/components under real plant conditions does not exist. Therefore, additional research and development works including large-scale tests for the direct substantiation of the operability of a new system may be needed individually for each advanced water cooled reactor concept. In particular, even the evolutionary plant designs may require confirmatory testing of some components and systems prior to commercial deployment.

Many experiments have been performed up to now with natural circulation systems/components used in new reactor concepts, and many of them have displayed a number of phenomena that are not adequately predicted by the existing computer codes. For example, it was found that gas stratification in the containment can significantly affect the efficiency of building condenser when the cooled H2-layer at the top displaces the steam. In advanced BWRs (e. g. the SWR-1000), vent pipes are therefore used connecting a position above the building condenser with the wet well. If heat transfer in the building condenser deteriorates because of steam shortage, the pressure in the containment increases and forces self-controlled venting of H2/steam mixture to the wet well [15].

The functional reliability of a natural circulation system depends essentially upon the way the natural physical phenomena operate in a particular system and the long-term effect of the environment on the system performance. It requires the identification and quantification of the uncertainties in the interaction between the phenomena, the immediate surroundings and the natural circulation system itself. In this respect, collection of existing and generation of new experimental data would provide information on influences on the functional reliability of the natural circulation systems.

Keeping in mind the above considerations, extensive analytical and experimental works are still needed as applied to new passive systems/components. The contents of these works and the relevant capabilities are discussed in Sections 3 and 4 of this report. The sharing of the relevant results obtained earlier and coordination of the further works on the international level would increase the value of their results and optimize the required manpower and financial resources.