For certain steam line break (SLB) analyses, the RETRAN-3D Oconee model predicts water carryout through the break. The amount of liquid carried from the system is dependent on the assumed break size, feedwater boundary conditions, and initial conditions. Although liquid carryout is potentially realistic during this scenario, there are no plant or test facility data on which to validate the amount of water carryout.

Because the steam generator tube stress evaluation is a safety related analysis, a conservative approach is required. Overcooling effects for an SLB transient are maximized by retaining as much steam generator liquid as possible to remove energy from the reactor coolant system. Because water carryout is nonconservative, and due to the lack of data, the analysis is performed assuming no liquid is carried from the break. This boundary condition is difficult to impose on a RETRAN-3D analysis since the amount of liquid entrained in the break flow is determined by internal code models, which the analyst cannot control directly.

This paper presents the methodology used to eliminate water carryout for these types of calculations. The methodology consists of a combination of special RETRAN-3D code modifications and model input changes.

In the second part of the paper, the results of an SLB analysis for the Oconee Nuclear Station employing the above methodology are presented. These analyses are done to compute the temperature differences between the steam generator tubes and the shell of the once-through steam generator. The temperature of the thin tubes decreases much faster than the temperature of the shell during an overcooling transient such as an SLB, resulting in tensile stresses that might lead to tube failures.

A number of break sizes were analyzed starting with a double-ended main SLB down to a small break of 0.0372 m2 (0.4 ft2). The sensitivity of the tube tensile stress to the assumed break size is presented.