Primary coolant circulation loop

We shall start this chapter with a short presentation of the different flow paths in our hydraulics model of the BWR with natural circulation. In the subsequent sections we have a more in-depth treatment of the individual flow paths. The system model of the hydraulics problem in a BWR with natural circulation is illustrated in Figure 5.1. The flow paths are assumed to have constant flow area, A_c, and equivalent hydraulic diameter, D_e.

$\begin{figure} % latex2html id marker 20016\rule{\textwidth}{0.2mm} \rule{0cm}... ...hefigure}\hspace{1em}Modeling of the different flow paths of a BWR.}\end{figure}$

The recirculation flow enters the core, $\framebox{1}$ , from the lower plenum, $\framebox{6}$ . Before the subcooled water enters the fuel assemblies it has to pass the core plate, which in forced-circulation BWRs holds the orifices which optimizes the flow split between the fuel assemblies. In a natural circulation BWR one cannot, in general, afford the appreciable pressure drop induced by orificing the core flow due to limitations in the natural circulation driving force. Therefore the loss coefficient associated with the core plate, K₀, is considerably smaller than normally encountered in BWR design.

From the bottom to the top of the core the water eventually starts to boil such that we at the top of the core have a saturated steam-water mixture with a quality around 15 %. From the top of the core the steam-water mixture enters the riser (or chimney), $\framebox{2}$ , which is a single circular tube with the diameter of the core. The riser enhances the natural circulation driving force.

Following the riser a steam separator, $\framebox{3}$ , separates the liquid from the vapor which is directed to the turbine. In practice, it is impossible to obtain a complete separation, ie the liquid will contain a little fraction of vapor bubbles and the vapor will contain small liquid droplets--these two effect are called carry under and carry over respectively.

The saturated liquid (+the small amount of vapor) returned by the steam separator assembly is mixed with the (highly) subcooled feedwater in flow path $\framebox{4}$ . In steady-state the feedwater mass flow rate corresponds to the total mass flow rate at the vapor outlet of the steam separator. To reduce thermally induced stresses and to improve the mixing characteristics it is customary BWR design to add the feedwater through a sparger (a circular perforated pipe which is placed internally in the flow path).

The combined liquid from the steam separator and the feedwater, the so-called recirculation water, now flows through the annular downcomer flow path, $\framebox{5}$ . From the downcomer the recirculation water enters the lower plenum flow path, $\framebox{6}$ . The flow in the lower plenum is complicated by the control rod guide tubes which are situated vertically in the lower plenum.

From the lower plenum the recirculation water again enters the core, $\framebox{1}$ .

As pointed out by many authors in the literature concerned with the hydraulics of the boiling water reactor it is the data of the external flow paths which have the largest uncertainty. If we disregard the steam separator, the solution (eg the recirculation mass flow rate) exhibits a very low sensitivity to a variation of the external flow path data. Unfortunately this is not the case with the steam separator since it will normally induce a pressure drop comparable to that of the core [20] (see also Table 15.2 p.

). It is however standard practice (when modeling forced-circulation BWRs) to utilize a very rough model of the steam separator. Inside the steam separator (in the stand pipes) the two-phase mixture undergoes a spinning movement which literally throws the liquid against the pipe wall. Since this kind of flow is 3-D in nature and has a very fine and complex structure it is virtually impossible to describe the behaviour of a steam separator on a firm theoretical basis. As a consequence we are going to model the steam separator with an equivalent length of pipe--an approach also taken by the authors of the NATBWR code [41]. In general estimation of flow path areas, hydraulic equivalent diameters and equivalent lengths for the external flow paths require a very subjective judgement and are consequently highly uncertain.

In addition to the pressure drops associated with the flow paths we have irreversible pressure drops which accompanies discontinuities in flow area when the fluid enters a new flow path--these pressure drops are frequently called singular pressure drops.

Putting the singular pressure drops aside for the moment (these pressure drops are considered in the last part of this chapter) we will now investigate the individual flow paths.

Readers not familiar with the notation used in connection with two-phase flows should consult section 6.4 (see p.

) before proceeding with the next sections.