Sodium spray fire analysis (II)

not registered  ; not registered

Sodium spray fire experiments using a 21 m vessel were Performed in order to validate the computer code SPRAY-III. The results of precalculations has been reported in a preceding report, Sodium Spray Fire Analysis. In this study, the validation of SPRAY-III was conducted using the experimental results and the accuracy of calculated results was discussed. The following results were obtained: (1)study under inert gas atmosphere. (a)The peak pressure and temperature obtained by the analysis overestimate the experimental results. But the calculated pressure rose monotonically till the termination of spray injection and it was different from the experimental peak which appeared during the injection. The rise rates of pressure and temperature in analyses were lower than those in experiments. (b)The calculated temperatures of the steel wall liner were only about 55 to 75 % of the experimental values at the end of sodium injection. But the comparison between analysis and experiment with regard to the temperature of mortar wall showed a fairly good agreement. (c)The averaged pool temperature at the end of sodium injection was found to be slightly lower in analysis than in experiment. The analysis shows that the heat transfer between the pool and the gases hardly affects the results during sodium injection. (2)Study under air atmosphere. (a)When all reaction products were assumed to be peroxide in the same manner as in the safety analysis of Monju, the increase rates of both the pressure and the temperature as well as their peak values were found to be underestimation. Hence to obtain conservative results in this 21 m vessel, an increase in the rate constant of reaction with oxygen (Vo from 300 to 1000 ft/sec) must be assumed. However, this value should be investigated in the near future for application to the safety analysis of Monju. (b)The comparison between analysis and experiment with regard to the temperature of the mortar showed a fairly good agreement. ..