During severe accident sequences hydrogen is produced, and together with steam it is released into the containment. For some accident sequences in BWRs this flow is injected into a water pool; in this case most fission products are contained in the water pool. If the steam/hydrogen flow is released into a dry containment the hydrogen is distributed in the containment and its distribution is influenced by hot and cold structures, condensation, sedimentation and the heat input by fission product decay. The knowledge about the hydrogen distribution is important for the assessment of whether hydrogen accumulates somewhere reaching concentrations with the potential for detonation processes.

Hydrogen detonation can be avoided either by pre-inertisation (as for all BWR containments), by post-inertisation or dilution or by reducing the amount of hydrogen using igniters or catalytic recombiners. It is evident that the mixing of different gas flows as well as the heat addition by recombiners and igniters create complex three-dimensional velocity and concentration fields in the containment volume; the effectiveness of recombiners and igniters are influenced by these fields.

Several conclusions can be drawn from the survey of experimental data and the identification

of needs; they are listed below:

a) The development and validation of CFD codes increased the requirements on the quality and details of experimental data. This has, e. g. led to a new instrumentation like tomography of two-phase flows to get knowledge about the spatial distribution of different fluids or phase. In addition, local turbulences on a micro scale should be modelled. New challenges result for the quality of experimental results from the required knowledge of detailed interfacial phenomena. The required knowledge as listed above should be available for steady state phenomena and transient flows.

b) For sequences with phase changes, especially for the condensation process, the influence of non-condensable gases on interfacial interactions, heat transfer and friction losses has to be investigated in more detail as compared with existing knowledge and with ongoing research.

c) For reactors using the natural circulation mode for core cooling, the neutronic-thermal hydraulic interactions are of vital importance because these might lead to core and flow instabilities. Instabilities with large amplitudes may influence stable operation.

d) A validation of models in computer codes should be based on different experiments in different facilities — if possible. Nevertheless the necessity to have well-based scaling laws to extrapolate from smaller scales to real plant sizes is of high importance. In addition, data from plant transients performed intentionally or unintentionally should be used more often for code validation.

e) There is a large amount of data from Separate-Effect tests and Integral tests selected systematically by CSNI for the purpose of code validation; see [Validation Matrices] for
the requirements on the selected data and the matrices. Nevertheless, some special tests focusing on NC-relevant phenomena and system behavior should be incorporated into these matrices. In addition, it should be evaluated whether further tests are needed to increase the confidence in the experimental database.

This safety function is ensured by protecting and maintaining the integrity of the potential radioactivity release barriers (fuel, reactor system boundary and containment). These barriers are passive components as themselves; in addition, several passive means are proposed in V — 407 and V-392 concepts for the protection of these barriers (some of them are reflected above). As the containment is the last and most important barrier, both these designs imply substantial improvement of the containment protection against different loads related to

design basis and severe accidents, and various passive systems are important part of this protection in V-407 and V-392 designs. This design decision is derived from the assumption that the active systems are more vulnerable to failures under conditions inside the containment during an accident.

In V-407 design, containment over-pressurization is avoided by passive containment cooling system (C-PHRS) as described above. To limit considerably the release of fission products beyond the containment, a permanent under-pressure is maintained in the inter-containment gap of the V-392 design. This safety function, one of the most important, is fulfilled by two systems: (1) an exhaust ventilation system equipped with a filtering plant with suction from the inter-containment gap and outlet into the stack; (2) a passive system of suction from the inter-containment gap. The first system is intended to control removal of steam-gas mixture from the inter-containment gap under accidents with total loss of power. The system is capable to remove at least 240 kg per hour that is equivalent to the inner containment leaks of 1.5% containment volume per 24 hours. The second system consists of lines connecting the inter-containment gap with the PHRS exhaust ducts, which are always in the hot state. This solution enables permanent removal and purification of inner containment leaks irrespective of the electricity supply and operator actions. According to estimations, the under-pressure is maintained at any point of the inter-containment gap with inner containment leaks up to 2.8% of containment volume per day (the design basis for the containment is 0.3%). The technical solution described above in combination with the systems for the containment pressure decrease (traditional spray system and new passive heat removal system) allows to give up the filtered venting system designed for V-392 in spite of this system follows the current requirements that filtered venting should not increase the risk of loosing the containment function and filtered venting is not required in the short term of a core melt accident.

Special systems and components are implemented in both new WWER designs to prevent hydrogen burning or explosion. For example, in V-392 design the hydrogen suppression system comprises passive catalytic hydrogen igniters based on an efficient high porosity cellular material. Each of 50 elements of this system is capable to oxidize about 30 grams of hydrogen per hour at its volumetric concentration 4%. This system prevents the explosive concentration of hydrogen even if 100% of the core Zr will be oxidized during an accident.

The governing equations for the steady state condition are obtained by dropping the time dependent terms. Also, by definition w (the non-dimensional flow rate) is unity at steady state. Therefore, the steady state momentum and energy equations can be written as:

The steady state solutions for the temperature of the various segments of the loop can be obtained from Eq. (13) (see Vijayan (1999)). Using these steady state solutions, the integral in

equation (12) can be calculated as (jd dZ = 1,for the loop shown in Fig. 1. Hence, the steady

state flow rate can be expressed as

Where C=(2/p)r and r = 1/(3-b). Thus, knowing the value of p and b the constants C and r in equation (14) can be estimated. For laminar flow (where p=64, and b=1) equation (14) can be rewritten as

Similarly, assuming Blassius friction factor correlation (p=0.316 and b=0.25) to be valid for turbulent flow we can obtain the following equation

In the transition region, one can expect a continuous change in the exponent of equation (14) from 0.5 to 0.364 as well as for the constant from 0.1768 to 1.96. It may be noted that both equations (15) and (16) are the exact solutions of equation (12) assuming the same friction factor correlation to be applicable to the entire loop. For closed loops, often the same friction factor correlation may not be applicable for the entire loop even if the loop is fully laminar or fully turbulent. An example is a fully laminar loop with part of the loop having rectangular cross section and remaining part having circular cross section. Hence, one has to keep in mind the assumptions made in deriving the relationship (14) while applying it.

The relationship expressed by Eq. (14) is derived for a rectangular loop with both the heater and cooler having the horizontal orientation. For other orientations also, it can be easily shown that the same relationship holds good if the loop height in the definition of the Grm is replaced with the centre-line elevation difference between the cooler and the heater, Dz (see

Vijayan et al. (2000)). The same is true for other loop geometries like the figure-of-eight loop used in Pressurised Heavy Water Reactors (PHWRs). For the figure-of-eight loop, the heater power used in the Grm is the total power of both heaters. The equations are also applicable for identical parallel-channel or parallel-loops systems if the parallel channels/loops are replaced by an equivalent path having the same hydraulic diameter and total flow area.

To reduce the computational effort only half of the complex but symmetric geometry of the test facility is simulated. According to the experience gained from the numerical analysis of SUCOS-2D, all structures have to be modeled in detail and a very fine grid is necessary for a good resolution of the thin boundary layers near walls and coolers. The mesh size of the grid changes from 8 mm to 1 mm. Ratios of the mesh sizes between two neighboring cells are less than or equal to 2. The 3d grid consists of 691,000 fluid cells and 68,000 structure cells. A first order upwind scheme is used to compute the convective fluxes of enthalpy and momentum. No turbulence model is used because the flow was laminar in the experiment.

The connecting tubes of the coolers, which are present in the horizontal side area are spatially recorded and modeled even if it is expected that they would have a minor influence on the natural convection than in the calculation of SUCOS-2D. The heat losses to the outside through the lateral walls are neglected. The active heat exchangers can be modeled by a heat exchanger model or by pre-setting a distribution of surface temperature or of heat flux. The calculation of SUCOS-2D showed that it is not necessary to simulate the coolers with the complex heat exchanger model. It is sufficient to give a distribution for the temperature on the surface between fluid and cooler. A linear distribution for the temperature is approximated by a step function prescribing three values for the vertical right coolers. For the horizontal coolers a constant value of temperature is sufficient because the difference of temperatures between inlet and outlet coolant water is less than 1 K. The prescribed values are determined by means of experimental data.

In former simulations for SUCOS-2D it was found that the heated copper plate needs special attention (Kuhn 1996). Even the developing circulation sense in the complete fluid domain is sensitive to the thermal boundary conditions used at the upper surface of the copper plate (Grotzbach et al. 1997). There, the problem of using an artificial Neumann or Dirichlet bound­ary condition was analyzed by calculating the heat conduction in the copper plate. 2d tests showed the surprising result that the copper plate does not ensure a constant heat flux to the fluid, but that it redistributes the heat horizontally in such a strong manner, that the heat flux

into the fluid varies along the plate surface by more than +/-50% of its mean value, Fig. 4. Thus, the thermal conduction in the heater plate is also calculated here. A 3d grid is used for the heated plate; the horizontal grid width distribution corresponds to the one of the fluid re­gion; 5 cells are used in the vertical direction with mesh sizes of 6 mm. The electrical heaters below the copper plate are simulated as a heat flux boundary condition with constant horizon­tal distribution.

The simulation was performed on a CRAY J916 with a memory need of 2.7 Gbytes. The tran­sient calculation was preceded by a steady state calculation to obtain an initial flow and tem­perature field for the transient calculation. The steady state calculation is stopped when an equilibrium in the changes of temperature and in the balance of the heat fluxes is nearly achieved. This happened after 4 h corresponding to 240 h of CPU-time. The transient calcula­tion is performed for a problem time of 227 s with a time step width of 1.0 s. This corresponds to 407 h CPU-time. The system of the pressure equations is solved by the iterative CRESOR method (Borgwaldt 1990), whereas the system of the enthalpy equations is solved by the itera­tive SOR method.

In two-phase flow, local convective heat transfer coefficients are evaluated from the channel local thermal hydraulic conditions and the heater rod surface temperature. In theory all the main two-phase flow patterns such as bubble flow, slug flow, churn flow and annular flow are likely to exhibit different heat transfer characteristics. It is generally recognized that more accurate prediction of flow boiling can be obtained by adopting a flow pattern specific method for each individual flow regime. However, practically all of the existing correlations correlate flow boiling data using a single equation covering all the different flow patterns. General boiling curve is used to define the wall heat transfer regimes (convective, pool boiling, flow boiling, film boiling, etc.) and the related convective heat transfer coefficient. In addition, distinction must be made for condensation and evaporation. Further, well-established models for the critical heat flux (CHF) are required for the calculation of pre-CHF and post-CHF wall heat transfer.

It is important to note that stability limits in natural circulation systems arise before (and as a prelude to) CHF or DNB. Indeed, conventional forced flow CHF and DNB correlations cannot be applied to natural circulation and parallel channel systems if either the loss coefficients are unknown or not reported, or the appropriate constant pressure drop was not maintained or achieved in the tests. Throttling the inlet flow to set a flow boundary condition artificially stabilizes the channel. In actual plants, it is well known that the plant maintains a constant pressure drop, by having multiple parallel channels and/or a controlled downcomer hydrostatic head.

Designs of pressurized systems limit the heat removal to that determined when there is no bulk boiling. The flow is always subcooled, and the heat exchange is by single-phase (liquid) flow in the heat exchanger. Heat removal in normal and accident conditions can be set by the convective heat removal by natural circulation.

We would like to establish the maximum or ultimate heat removal in a natural circulation design where there is a known elevation difference between the heat source (core) and sink (HX). The maximum power limit is when the heat generated is completely removed by the heat exchanger loop, and the HX outlet temperature is close or equal to the secondary side (boiling) temperature. The turbine stop valve (design) pressure of course sets the secondary

side temperature. Thus, there is a relation between the core maximum (saturation or subcooled) temperature, and hence the power, and the secondary pressure.

For the purposes of the maximum design output evaluation, we define the maximum core outlet temperature as the saturation temperature at the primary pressure. Thus the ultimate limit is taken as bulk boiling at the core exit and not conventional DNB. We consider the two — phase (boiling) limit later and use results available from the literature (Duffey and Sursock, 1987); (Duffey and Rohatgi, 1996).

The overall loop flow W, in a natural-circulation system with a thermally expandable fluid, utilizing the Boussinesq approximation, Ap= pfiAT, for the expansion coefficient, Д is given by the well-known result

Here A is the flow area, K the loop loss coefficient, AZ the effective available driving head between the heat source and sink, Q the power, g acceleration due to gravity, pe the liquid density and cp the heat capacity.

Since the two — phase flow rate W24> in the loop is given very nearly by:

or

1/3

‘P

q 2 hg b fgH

where Ф = pt/(pt — pg) and hfg is the latent heat and Cp /hfgb is a dimensionless evaporation number.

Thus we expect to find that most major loop and system parameters have a relatively weak (one-third power) influence on the flowrate.

Seven commercial NPP systems and three ITF, not used for setting up the database presented in Fig. 2, are considered for demonstrating the use of the NCFM, Ref. [3]. Main characteristics of the NPP and of the ITF can be drawn from Tables 2 and 3, respectively.

TABLE II. RELEVANT CHARACTERISTICS OF NPP CONSIDERED FOR THE APPLICATION OF THE NCFM

TABLE III. RELEVANT CHARACTERISTICS OF ITF CONSIDERED FOR THE APPLICATION OF THE NCFM

Reactors 1 to 4 (Table II) have been built and are in operation. Reactors 5 to 7 are in a more or less advanced design stage. The geometric layout of primary systems for reactors 1, 2, 5, 6 and 7 is similar to the geometric layout of ITF originating the database for the NCFM. However, differences are present in the relative elevations between core and SG. Reactors 3 and 4 are equipped with OTSG and HTSG, respectively. So the geometric layout of the primary system is different from the geometric layout of ITF originating the database for NCFM.

Pactel and RD14M (Table III) are experimental simulators of WWER-440 and CANDU NPP, respectively. Their geometric layout is different from those of a PWR. In the case of WWER — 440, six loops equipped with HTSG are connected to the vessel, though only three are simulated in Pactel. Horizontal core configuration characterizes the CANDU design, that otherwise is equipped with UTSG.

A comparison has been made between measured (case of ITF) and calculated (case of NPP) system behaviours during NC and the data that characterize the NCFM. To this aim, code calculations assuming stepwise draining of primary system fluid mass inventory have been performed (case of NPP) and relevant NC experimental data are utilized (case of ITF).

In the case of NPP, the qualification level of the adopted code and nodalisation affects the calculated NCP. Furthermore, in the case of new generation ‘passive’ safety reactors 6 and 7, the emergency loops connected with the primary system are assumed to come into operation once the coolant draining process is initiated.

Calculated or measured transient scenarios reflect the NC flow regimes identified in Fig. 1, Ref. [3]. The RCNC regime is not achieved in the systems equipped with HTSG and OTSG and is not evident from the RD14M database. The SCNC regime is also not evident in all the calculations or available experimental databases. Mostly SPNC is calculated in the NPP 6 and 7.

Significant results are shown in Figs 4 to 6. The following observations apply:

• NCP of UTSG equipped PWR and of WWER-1000 is qualitatively similar (Fig. 4). Therefore the last generation of HTSG WWER shows a reasonable NCP. The good performance of the PWR-2 can be noted (low power NPP equipped with four UTSG).

• The OTSG equipped PWR show an ‘early’ NC flowrate decrease and an early stop of NC due to void formation in the ‘candy-cane’ and the rising part of the hot leg (Fig. 4).

• NC flowrate in AP-600, as expected, is not affected by draining because of liquid mass supplied by the ‘passive’ emergency cooling loops (Fig. 5). This behaviour does not show up in the case of EP-1000 presumably due to lack of qualification of the adopted code model.

• The WWER-440 simulator (Pactel) exhibits a decrease of the NC flowrate at relatively high mass inventories of the primary loop. The presence of the hot leg loop seal is at the origin of a partial flow stagnation (Fig. 6). Removal of the hot leg loop seal is effective in improving NCP as shown by the calculated WWER-1000 transient. The consideration of a passive system also improves the NCP of the Pactel.

• The CANDU simulator exhibits two different behaviours depending upon the considered experiment (Fig. 6). This shows the need of a deeper investigation before drawing conclusions. It may be noted that larger driving forces characterize CANDU systems for NC compared with PWR systems, owing to the larger distance of heat source and sink. However, larger pressure drops are also expected owing to the longer core and to the presence of small equivalent diameter pipes at core inlet and outlet.

Any attempt to judge the results, i. e. the NCP of involved NPP and ITF, should consider the quality of the used databases (DB). The result of a two step evaluation can be found in Table IV. The second column considers the demonstration of the quality of the starting DB. If ‘N’ appears in the second column, the evaluation in the third column is not meaningful. The third column considers the quality demonstration of the used DB, e. g. the nodalisation in the cases of code use and, finally, the NCP judgement (more details are provided in Ref. [3]).

The ‘NS’ mark for PWR-3 is due to the poor NCP also at high values of the RM inventory caused by the long vertical hot leg. Quality of data is assumed to be suitable. The ‘N’ in the case of EP-1000 (third column) is related to the available nodalisation. The dash (-) in the case of RD14M comes from the contradictory DB available at the moment.

TABLE IV. SUMMARY OF NCP EVALUATION FOR THE CONSIDERED NPP AND ITF

A separate paper is given in this meeting to describe the passive safety experiments in more detail. Here is only a short overview of the experiments.

1.1. Experiments with CMT and two PBL’s

The main objective of the tests was the overall simulation of the PSIS behaviour. The experiments demonstrated the capability of PACTEL for the simulation of PSIS’s [2, 3, 4, 5]. The data did not provide very detailed information about phenomena in the CMT. The instrumentation of the CMT in the first test series was limited. The tests were partially aimed for investigation of possibilities to run passive safety injection tests on PACTEL. The experience gathered during this first series was then used in the specification of the test parameters and instrumentation of the second series.

1.1.1. The first series ofpassive safety injection experiments (GDE-01 through GDE-05)

The first experiment series included five SBLOCA tests with break in one hot leg of PACTEL. Three break sizes were used. Three tests included secondary system depressurization as an accident management measure. In all tests, only PSIS provided ECC water to the core. The primary pressure used in the tests was lower than the nominal operation pressure of PACTEL. Maximum operation pressure of the passive accumulators determined the upper limit of the experiment pressure (3.8 MPa).

Although some problems with condensation in the CMT occurred no core heat-up was detected. It should be mentioned here that the CMT used in the test was very large, about twice as large as the scaled volume of four accumulators of the reference plant. On the other hand, the volume of the CMT is about the same as the volume of the rest of the loop. All the tests in this first series were terminated before the CMT was totally empty.

The second experiment series included four SBLOCA tests with break in one cold leg of PACTEL. Two break sizes were used. The tests also included studies of primary system depressurization and reproducibility of the phenomena. In all tests, only the CMT provided ECC water to the primary loop.

The primary pressure used in the tests was again lower than the nominal operation pressure of PACTEL. The maximum operation pressure of the passive accumulator limited the maximum experiment pressure to 3.8 MPa. Since one loop of PACTEL was equipped with a different steam generator model the tests run with only two loops in operation. There was still water in the CMT when the tests were terminated. During the rapid condensation period, water level in the core simulator dropped close to the top of the core. However, also in these test no core heat-up occurred.

The NOKO facility has also been used to study the effectiveness of systems removing the decay heat from the containment.

3 4 5

Level over Drain Line Connection [m]

In Fig. 8 the configuration of the building condenser within the condenser tank is shown.

In Fig. 9 test results showing the influence of non-condensables on the energy transport from the building condenser to the environment are shown.