The Marviken Full Scale Critical Flow Tests (CFT) were conducted between mid-1977 and December 1979 as a multinational project at the Marviken Power Station. The 27 CFT experiments, together with the test procedures, equipment, and evaluation techniques are presented in a set of 35 reports. Data describing the nozzle inlet stagnation conditions and mass fluxes are presented.
The tests were conducted by discharging water and steam-water mixtures from a full sized reactor vessel through a large diameter discharge pipe that supplied the flow to the test nozzle. The nine test nozzle geometries all had rounded entrances followed by a nominally 200, 300 or 500 mm constant diameter cross-section. The nozzles ranged in length from 166 to 1809 mm.
Most tests were conducted with a nominal initial steam dome pressure of 5 MPa (4 MPa for one test) and with the water initially subcoo1ed between 50 deg.C and 1 deg.C (with respect to the steam dome pressure). The water oxygen content was approximately the same for all tests, but it was decreased by a factor of 140 with respect to a reference test in order to determine the effect of the water air content on the critical mass flow rate.
The vessel, discharge pipe and nozzle were instrumented to determine the test behaviour and to provide a basis for evaluating the stagnation conditions and mass fluxes at the nozzle inlet. The instrumentation readings were recorded using a pulse code modulation system. Data error limits were evaluated for all measurements.
Using the recorded measurements, changes in the fluid state which occurred while the test fluid flowed from the instrumentation station to the nozzle inlet were examined. The resulting nozzle inlet conditions are presented. The most accurate rates of mass flow through the nozzle were obtained using a pitot-static rake during the subcooled portion of the experiments and by tracing the liquid mass history during the saturated portion of the experiments. The peak mass flow rate measured during the CFT program was 14 Mg/s.
Nozzle pressure profiles showed a pressure recovery zone just downstream of the nozzle inlet during the subcooled test phase and usually a steadily- decreasing pressure during the saturated test phase.
Examination of the subcooled data recorded for the nine nozzle geometries indicates that the different nozzle mass fluxes lie within the CFT program error limits at L/D (length to diameter) ratios greater than 1. 5. However, the mass flux behaviour seems to be dependent on the diameter as well as the length at L/D's less than 1.5. A more clear dependence on length and diameter seems to apply for saturated flow with zero quality. The decrease in the
initial water air content had no observable effect on the nozzle mass flux histories.
The organizations represented were from Denmark, Germany, Finland, France, the Netherlands, Norway, Sweden and the United States.
The official documentation of the project is given in the reports and appendices, the titles of which are listed on the back cover.