Experimental study on the air/water counter-current flow limitation in a model of the hot leg of a pressurized water reactor

An experimental investigation on the air/water counter-current two-phase flow in a horizontal rectangular channel connected to an inclined riser has been conducted. This test-section representing a model of the hot leg of a pressurized water reactor is mounted between two separators in a pressurized experimental vessel. The cross-section and length of the horizontal part of the test-section are (0.25 m × 0.05 m) and 2.59 m, respectively, whereas the inclination angle of the riser is 50°. The flow was captured by a high-speed camera in the bended region of the hot leg, delivering a detailed view of the stratified interface as well as of dispersed structures like bubbles and droplets. Countercurrent flow limitation (CCFL), or the onset of flooding, was found by analyzing the water levels measured in the separators. The counter-current flow limitation is defined as the maximum air mass flow rate at which the discharged water mass flow rate is equal to the inlet water mass flow rate.From the high-speed observations it was found that the initiation of flooding coincides with the formation of slug flow. Furthermore, a hysteresis was noticed between flooding and deflooding. The CCFL data was compared with similar experiments and empirical correlations available in the literature. Therefore, the Wallis-parameter was calculated for the rectangular cross-sections by using the channel height as length, instead of the diameter. The agreement of the CCFL curve is good, but the zero liquid penetration was found at lower values of the Wallis parameter than in most of the previous work. This deviation can be attributed to the special rectangular geometry of the hot leg model of FZD, since the other investigations were done for pipes.     

Erratum to “Experimental study on the air/water counter-current flow limitation in a model of the hot leg of a pressurized water reactor”

This paper replaces the paper published in the journal by Deendarlianto et al. (2008). Because of an error in the implementation of the air flow meter some of the data given by Deendarlianto et al. (2008) are wrong. They are corrected within the present paper. The general results and conclusions remain unchanged.An experimental investigation on the air/water counter-current two-phase flow in a horizontal rectangular channel connected to an inclined riser has been conducted. This test-section representing a model of the hot leg of a pressurized water reactor is mounted between two separators in a pressurized experimental vessel. The cross-section and length of the horizontal part of the test-section are (0.25 m × 0.05 m) and 2.59 m, respectively, whereas the inclination angle of the riser is 50°. The flow was captured by a high speed camera in the bended region of the hot leg, delivering a detailed view of the stratified interface as well as of dispersed structures like bubbles and droplets. Counter-current flow limitation (CCFL), or the onset of flooding, was found by analyzing the water levels measured in the separators. The counter-current flow limitation is defined as the maximum air mass flow rate at which the discharged water mass flow rate is equal to the inlet water mass flow rate.From the high-speed observations it was found that the initiation of flooding coincides with the formation of slug flow. Furthermore, a slight hysteresis was noticed between flooding and deflooding. The CCFL data was compared with similar experiments and empirical correlations available in the literature. Therefore, the Wallis-parameter was calculated for the rectangular cross-sections by using the channel height as length, instead of the diameter. The agreement of the CCFL curve is good, but the zero liquid penetration was found at lower values of the Wallis parameter than in most of the previous work. This deviation can be attributed to the special rectangular geometry of the hot leg model of HZDR, since the other investigations were done for pipes.     

Numerical simulation using CFD software of countercurrent gas-liquid flow in a PWR hot leg under reflux condensation

In order to improve the countercurrent flow model of a transient analysis code, countercurrent air-water tests were previously conducted using a 1/15 scale model of the PWR hot leg and numerical simulations of the tests were carried out using the two-fluid model implemented in the CFD software FLUENT 6.3.26. The predicted flow patterns and CCFL characteristics agreed well with the experimental data. However, the validation of the interfacial drag correlation used in the two-fluid model was still insufficient, especially regarding the applicability to actual PWR conditions. In this study, we measured water levels and wave heights in the 1/15 scale setup to understand the characteristics of the interfacial drag, and we considered a relationship between the wave height and the interfacial drag coefficient. Numerical simulations to examine the effects of cell size and interfacial drag correlations on numerical predictions were conducted under PWR plant conditions. Wave heights strongly related with the water level and interfacial drag coefficient, which indicates that the interfacial drag force mainly consists of form drag. The cell size affected the gas velocity at the onset of flooding in the process of increasing gas flow rate. The gas volumetric fluxes at CCFL predicted using fine cells were higher than those using normal cells. On the other hand, the cell size did not have a significant influence on the process of decreasing gas flow rate. The predictions for the PWR condition using a reference set of interfacial drag correlations agreed well with the Upper Plenum Test Facility data of the PWR scale experiment in the region of medium gas volumetric fluxes. The reference interfacial drag correlations employed in this study can be applied to the PWR conditions.     

Correlation for countercurrent flow limitation in a PWR hot leg

Numerical simulations were done to evaluate countercurrent flow limitation (CCFL) characteristics in a pressurized water reactor (PWR) hot leg with the diameter of 750 mm by using a volume of fluid (VOF) method implemented in the CFD software, FLUENT6.3.26. The calculated CCFL characteristics agreed well with known values including the UPTF data at 1.5 MPa. Sensitivity analyses for system pressures up to 8 MPa showed that the calculated CCFL characteristics in the Wallis diagram were slightly mitigated from 0.1 MPa to 1.5 MPa with increasing system pressure, but they did not change from 1.5 MPa to 8MPa. Using the CCFLs calculated in this study and values measured under air–water and steam–water conditions, a CCFL correlation and its uncertainty were derived.     

Condensation experiments for counter-current flow limitation in an inverted U-tube

In this study, we measured counter-current flow limitation (CCFL) characteristics in an inverted U-tube (18.4 mm diameter and 1.0 m straight-part length) simulating steam generator (SG) U-tubes under conditions of steam condensation at pressures of 0.1–0.14 MPa. Differential pressure ΔP between the top of the inverted U-tube and the lower tank was measured, and the flow patterns wave estimated by comparing the waveforms of ΔP with those in air–water experiments. As a result, we classified the flow patterns under CCFL conditions into CCFL-P, CCFL-L and CCFL-T. The falling water flow rate under CCFL conditions slightly increased as the pressure increased and the cooling water temperature decreased (subcooling of cooling water increased). In the case of CCFL-L, CCFL characteristics in the inverted U-tube were between those in air–water and saturated steam–water experiments at 0.1 MPa. Furthermore, we derived a Wallis type CCFL correlation and its uncertainty from CCFL data, including previously measured data, i.e., J*^1/2_G + 0.88J_L*^1/2 = 0.76 ± 0.05.     

Study of countercurrent flow limitation in a horizontal pipe connected to an inclined one

Experiments with air and water in small hot leg reproductions were carried out aiming to acquire a better understanding of the countercurrent flow limitation (CCFL) in this geometry. The effects of various geometrical parameters of the test section and of the inlet water flow rate on the onset of flooding, on the partial delivery of water and on the zero liquid penetration were investigated. It was observed that while the onset of flooding is affected by the inlet water flow rate, the zero liquid penetration is independent of this flow rate. The results with partial delivery showed that, for a fixed air velocity, an increase in the horizontal length, or in the inclined length, of the flow channel leads to an increase of the water carried over by the air. On the other hand, in pipes with larger diameters the drag of the water is smaller. The experimental results showed small differences in the results for tests with inclination of the riser lower than 90°. For an inclination equal to 90°, the water carried by the air tends to be lower than in the others angles for a fixed air velocity. The study led also to a new correlation for the flooding.     

Counter-current flow limitations during hot leg injection in pressurized water reactors

A one-dimensional model is presented to predict counter-current flow limitations during hot leg injection in pressurized water reactors. Different from previous models, it may also be applied in case of high Froude numbers of the liquid flow, such as to be expected in the case of emergency coolant injection through the hot leg. The model has been verified with an extensive experimental program performed in the WENKA test facility at the Forschungszentrum Karlsruhe. Typical flow regimes were investigated for a wide range of flow conditions, simulated with air and water at ambient pressure and temperature, in a simplified Pressurized Water Reactor (PWR) hot leg geometry. Depending on the water and air flow rates, flow phenomena such as a hydraulic jump and flow reversal were experimentally observed. The theoretical model shows that not only the nondimensional superficial velocities of liquid and gas, but also the Froude number of the liquid at the injection point and the Reynolds number of the gas play an important role for the prediction of flow reversal. In case of a high liquid inlet Froude number, a flow reversal could only be observed if the liquid flow became locally subcritical, i.e. if a hydraulic jump occurred in the channel. The flow reversal is predicted by the presented model with good accuracy.     

Experimental study of gas–liquid two-phase flow through packed bed under natural circulation conditions

Dry-out phenomena in packed beds or porous media may cause a significant digression of cooling/reaction performance in heat transfer/chemical reactor systems. One of the phenomena responsible for the dry-out in packed beds is known as the counter-current flow limitation (CCFL). In order to investigate the CCFL phenomena induced by gas–liquid two-phase flow in packed beds inside a pool, a natural circulation packed bed test facility was designed and constructed. A total of 27 experimental conditions covering various packing media sizes (sphere diameters: 3.0, 6.4 and 9.5 mm), packed bed heights (15, 35 and 50 cm) and water level heights (1.0, 1.5 and 2.0 m) were tested to examine the CCFL criteria with adiabatic air–water two-phase flow under natural circulation conditions. Both CCFL and flow reversal phenomena were observed, and the experimental data including instantaneous and time-averaged void fraction, differential pressure and superficial gas–liquid velocities were collected. The CCFL criteria were determined when periodical oscillations of void fraction and differential pressure appear. In addition, the Wallis correlation for CCFL was utilized for data analysis, and the Wallis coefficient, C, was determined experimentally from the packed bed CCFL tests. Compared to the existing data-sets in literature, the higher C values obtained in the present experiment suggest a possibly higher dry-out heat flux for natural circulation debris systems, which may be due to the water supply from both top and bottom surfaces of the packed beds. Considering the effects of bed height and hydraulic diameter of the packing media, a newly developed model for the Wallis coefficient, C, under natural circulation CCFL is presented. The present model can predict the experimental data with an averaged absolute error of ±7.9%.     

Assessment of RELAP5/MOD3.2.2γ against flooding database in horizontal-to-inclined pipes

A total of 356 experimental data for the onset of flooding are compiled for the data bank and used for the assessment of RELAP5/MOD3.2.2γ predictions of counter-current flow limitation (CCFL) in horizontal-to-inclined pipes simulating a PWR hot leg. RELAP5 calculations show that higher gas flow rates are required to initiate the flooding compared with the experimental data if the L/D is as low as that of the hot legs of typical PWRs. Based on the present data bank, the new CCFL correlation is derived, which shows the L/D effect. The present correlation agrees well with the database within the prediction error, 8.7% and it is implemented into the RELAP5 and validated against the data bank. The predictions of the flooding limit by the modified version lie well on the applied CCFL curve if the L/D is lower than 22, which is the case of the hot legs of typical PWRs.     

One-dimensional three-field model of condensation in horizontal countercurrent flow with supercritical liquid velocity

A one-dimensional three-field model was developed to predict the flow of liquid and vapor that results from countercurrent flow of water injected into the hot leg of a PWR and the oncoming steam flowing from the upper plenum. The model solves the conservation equations for mass, momentum, and energy in a continuous-vapor field, a continuous-liquid field, and a dispersed-liquid (entrained-droplet) field. Single-effect experiments performed in the upper plenum test facility (UPTF) of the former SIEMENS KWU (now AREVA) at Mannheim, Germany, were used to validate the countercurrent flow limitation (CCFL) model in case of emergency core cooling water injection into the hot legs. Subcooled water and saturated steam flowed countercurrent in a horizontal pipe with an inside diameter of 0.75 m. The flow of injected water was varied from 150 kg/s to 400 kg/s, and the flow of steam varied from 13 kg/s to 178 kg/s. The subcooling of the liquid ranged from 0 K to 104 K. The velocity of the water at the injection point was supercritical (greater than the celerity of a gravity wave) for all the experiments. The three-field model was successfully used to predict the experimental data, and the results from the model provide insight into the mechanisms that influence the flows of liquid and vapor during countercurrent flow in a hot leg. When the injected water was saturated and the flow of steam was small, all or most of the injected water flowed to the upper plenum. Because the velocity of the liquid remained supercritical, entrainment of droplets was suppressed. When the injected water was saturated and the flow of steam was large, the interfacial shear stress on the continuous liquid caused the velocity in the liquid to become subcritical, resulting in a hydraulic jump. Entrainment ensued, and the flow of liquid to the end of the hot leg was greatly reduced.The influence of condensation on the transition from supercritical to subcritical flow as observed in the experimental data is also predicted with the three-field model. When the injected water was subcooled, condensation on the flow of continuous liquid caused a reduction in the flow of vapor and, consequently, a reduction in the interfacial shear stress. Therefore, the flow of liquid remained supercritical to the end of the hot leg at the upper plenum. The entire flow of injected water flowed to the end of the hot leg at higher flows of steam when the injected water was subcooled than when it was saturated. When the flow of vapor was large enough to cause a hydraulic jump in the subcooled liquid, the rate of entrained droplets was greatly increased. The interfacial surface area of the droplets was several orders of magnitude greater than for the continuous-liquid field, and condensation rate on the droplet field was also several orders of magnitude greater. When the flow of vapor from the upper plenum was at its greatest, most of the flow in the continuous liquid was entrained before reaching the upper plenum. The large flow of subcooled droplets caused three-quarters of the steam to condense.     

Development of a critical heat flux mapping method for fin-structured surfaces

In this study, we performed critical heat flux (CHF) experiments using structured surfaces to validate the parameter effects and understand their physical meanings. Experimental results showed that the CHF has a peak value as the fin geometry changes. Fins with height of 0.5 mm produced the largest CHF, 1.7 MW/m², and fins longer than 2 mm reduced the CHF values. To explain the results, a CHF mapping method was developed describing the liquid supply-side and demand-side limits. The liquid demand-side limit is governed by the heat removal capability, mainly the nucleate boiling, calculated using the hot spot model. We consider three liquid supply-side limits restricting the liquid supply to the heating surface: capillary limit and counter-current flow limitations (CCFLs). The capillary limit is determined by balancing the capillary pressure and viscous dissipation in the liquid film on the fin side. The CCFL in the structure is calculated using the Wallis correlation and the CCFL in the free volume limits the liquid downward flow by the vapor jetting from the heating surface. The CHF map for our experimental results successfully describes the CHF trend of the structured surfaces. As a result, we concluded that CHF mapping method is an effective means of explaining CHF in pool boiling.     

Prediction of counter-current flow limitation at hot leg pipe during a small-break LOCA

The possibility of hot leg flooding during reflux condensation cooling after a small-break loss-of-coolant accident in a nuclear power plant is evaluated. The vapor and liquid velocities in hot leg and steam generator tubes are calculated during reflux condensation cooling with the accident scenarios of three typical break sizes, 0.13, 1.02 and 10.19% cold leg break. The effect of initial water level to counter-current flow limitation is taken into account. It is predicted that the hot leg flooding is precluded when all steam generators are available for heat removal. It is also shown that both hot leg flooding and SG flooding are possible under the operation of one steam generator. Therefore, it can be said that the occurrence of hot leg flooding under reflux condensation cooling is possible when the number of steam generators available for heat removal is limited.     

Effect of steam condensation on countercurrent flow limiting in nearly horizontal two-phase flow

An analytical model that includes the steam condensation effect has been derived and a parametric study has been performed. In addition, a series of experiments were performed and a total of 34 experimental data for the onset of countercurrent flow limiting (CCFL) in nearly horizontal countercurrent two-phase flow have been obtained for various flow rates of water. Comparisons of the present CCFL data with slug formation models show that the agreement between the present as well as the existing model and the data is about the same. However, the deviation between Taitel and Dukler’s model predictions and the data is the largest when j_f<0.04 m s⁻¹. A parametric study of the effect of condensation using the present model shows that, when all local conditions are similar, the model predicted local gas velocities that cause the onset of flooding are slightly lower when condensation occurred. Based on the visual observation and the evaluation of the present work, it has been concluded that the criterion derived for the onset of slug flow can be directly used to predict the onset of inner flooding in nearly horizontal two-phase flow within the experimental ranges of the present work.     

Flooding in the pressurizer surge line of AP600 plant and analyses of APEX data

For the passive AP600 plant, the three stages of ADS (automatic depressurization system) valves are attached to the top of pressurizer. The existence of these valves makes liquid flow into and out of the pressurizer an important part of the dynamics during a small break loss-of-coolant accident. In this paper, counter-current flow limit (CCFL) in the surge line was analyzed. Specifically, CCFL in vertical piping, in slightly inclined horizontal piping, and in horizontal and vertical elbows were compared. The CCFL in the vertical section of the surge line was found to be the most limiting section. That is, the vertical CCFL controls the pressurizer liquid drain rate. This conclusion was tested and verified by comparing the predicted vertical CCFL against the counter-current flow states in the surge line, observed in small break LOCA tests conducted at the AP600 scaled test facility (APEX).     

Experimental investigation and CFD validation of countercurrent flow limitation (CCFL) in a large-diameter PWR hot-leg geometry

ABSTRACT

Countercurrent flow limitation (CCFL) is a phenomenon that consists of several flow patterns occurring simultaneously which produces a complex gas/liquid interface and interfacial momentum transfer, thus making it one of the most challenging two-phase flow configurations for computational fluid dynamics (CFD) validation. Numerous experimental investigations have been carried out in recent years regarding this, but most of those investigations were performed in small-diameter pipes or in non-pipe geometries (rectangular cross sections). A review of these experimental investigations has shown that the scale and geometry of the test section has a large impact upon the onset and characteristics of the CCFL. In order to provide a better understanding of this phenomenon in an actual pressurized water reactor (PWR) hot-leg geometry at a relatively large-diameter and scale, a test facility with a ～1/3.9 scale and a 190 mm inner diameter was constructed. Experiments were carried out at atmospheric pressure using water and air. High-speed recording was used to acquire high-quality images of the air/water interface. CCFL mechanisms, flow patterns, and the limits of the onset of CCFL and deflooding were experimentally identified. CFD simulations of two representative cases were carried out and assessed against experimental results. The analysis of the CFD simulations has provided insights into the improvements necessary for the accurate simulation of CCFL in large-scale geometries.     

Effects of fluid properties on CCFL characteristics at a vertical pipe lower end

The purpose of this study is to derive a counter-current flow limitation (CCFL) correlation and evaluate its uncertainty for steam generator (SG) U-tubes in a pressurized water reactor (PWR). Experiments were conducted to evaluate effects of the liquid viscosity on CCFL characteristics using air–40 wt% or air–60 wt% glycerol water solution and saturated steam–water at atmospheric pressure with vertical pipes simulating the lower part of the SG U-tubes. The steam–water experiments confirmed that CCFL characteristics could be expressed in terms of the Wallis parameters (J_G* and J_L*) for the pipe diameters of D = 14, 20, and 27 mm. A CCFL correlation was derived using the ratio μ_G/μ_L of the viscosities of the gas and liquid phases, μ_G and μ_L, as a correction term representing effects of fluid properties, where J_G^*1/2(μ_G/μ_L)^?0.07 was expressed by a cubic function of J_L^*1/2(μ_G/μ_L)^0.1. In the correlation, the constant C indicating the value of J_G^*1/2(μ_G/μ_L)^?0.07 at J_L* = 0 was (1.04 ± 0.05), and this uncertainty of ±0.05 would cover most of the previous experimental data including the ROSA-IV/LSTF data at 1, 3, and 7 MPa.     

反应堆压力容器下降段水-蒸汽CCFL实验与模型研究

为探究反应堆压力容器下降段在喷放末期冷段安注过程中的水-蒸汽逆流特性,建立下降段逆向流动限制(CCFL)模型,开展了基于压力容器模化本体的下降段CCFL实验研究以及建模分析。通过实验研究获得了不同入口安注水流量、安注水过冷度、堆芯蒸汽流量等条件下的下降段环腔内的安注特性数据,并基于实验数据进行了CCFL建模分析。结果表明,开始发生CCFL的蒸汽无量纲流速与入口安注水无量纲流速呈现正相关,基于无量纲流速建立的模型斜率与入口安注水无量纲流速呈现高度指数关联。本文建立了适用于从不发生CCFL至不完全CCFL,再到完全CCFL的下降段水-蒸汽气液逆流全过程预测模型。     

Two-phase countercurrent flow in a model of a pressurized water reactor hot leg

The onset of flooding or countercurrent flow limitation (CCFL) determines the maximum rate at which one phase can flow countercurrently to another phase. In the present study, the experimental data of the CCFL for gas and liquid in a horizontal pipe with a bend are investigated. The different mechanisms that lead to flooding and that are dependent on the liquid flow rate are observed. For low and intermediate liquid flow rates, the onset of flooding appears simultaneously with the slugging of unstable waves that are formed at the crest of the hydraulic jump. At low liquid flow rates, slugging appears close to the bend; at higher liquid flow rates, it appears far away from the bend, in the horizontal section. For high liquid flow rates, no hydraulic jump is observed, and flooding occurs as a result of slug formation at the end of the horizontal pipe. The effects of the inclination angle of the bends, the liquid inlet conditions and the length of the horizontal pipes are of significance for the onset of flooding. A mathematical model of Ardron and Banerjee is modified to predict the onset of flooding. Flooding curves calculated by this model are compared with present experimental data and those of other researchers. The predictions of the onset of flooding as a function of the length-to-diameter ratio are in reasonable agreement with the experimental data.     

Air—water model studies of cocurrent flow into and along a PWR hot leg to the steam generator

An experimental study is made of the cocurrent flow of air and water at atmospheric pressure from a reactor vessel into and along an approximately ninth-scale replica of the Sizewell ‘B’ PWR hot leg to the steam generator. A flow regime map of conditions in the hot leg is presented.The water interface level in the reactor vessel as a function of the flowrates is in agreement with a recent theory developed by the author. The same theory predicts the level in the hot leg when discharging two phases through a horizontal break and is in agreement with the results of other workers on this subject for the discharge of air and water up to a pressure of 5 bar and of steam and water up to a pressure of 62 bar.Results on the water level in the hot leg are correlated empirically but, for lower flowrates, the results are in approximate agreement with a theory for the onset of flooding.     

Development of Interfacial Friction Model for Two-Fluid Model Code against Countercurrent Gas-Liquid Flow Limitation in PWR Hot Leg

An interfacial friction model for two-fluid model code has been developed for the counter- current gas-liquid flow limitation at hot leg in a pressurized water reactor. Firstly, using a typical two-fluid model code TRAC-PF1/MOD1, we analyzed whether the interfacial friction model under countercurrent stratified flow by Ohnuki et al., which has been verified with an envelope model at steady state, functions well for the dynamic calculation with the two-fluid model code or not. It was found from the analyses that the model by Ohnuki et al. should be combined with a suitable interfacial friction model for the slug flow regime in hot leg. Based on flow observation in a small scale air-water experiment, models at the bend of hot leg and in the roll wave regime in the horizontal flow path of hot leg were newly developed as the model in the slug flow regime and the slug flow model was combined with the model by Ohnuki et al., The validity of the present model was confirmed with the data under various conditions of scales, pressures and fluid combinations (inner diameter: 0.025~0.75m, pressure: 0.1~7.1 MPa and air-water or steam-water).