Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 34, No. 8, p.

783-791 (August 1997)

Measurement
Flow

System
Characteristics

of Bubbly
of Bubbly

Flow

Using

Ultrasonic
Flow

Velocity Profile Monitor and Video Data Processing Unit, (II)

Countercurrent
Masanori ARITOMI*,t, Shirong ZHOU*, Makoto NAKAJIMA**, and Michitsugu MORI**** Yasushi TAKEDA***

* Research Laboratory for Nuclear Reactors , Tokyo Institute **Mitsubishi Heavy Industr y Company *** Paul Scherrer Institute **** Tokyo Electric Power Company (Received December 25, 1996)

of Technology

The authors have developed a measurement system which is composed of an ultrasonic velocity profile monitor and a video data processing unit in order to clarify its multi-dimensional flow characteristics in bubbly flows and to offer a data base to validate numerical codes for multi-dimensional two-phase flow. In this paper, the measurement system was applied for bubbly countercurrent flows in a vertical rectangular channel. At first, both bubble and water velocity profiles and void fraction profiles in the channel were investigated statistically. Next, turbulence intensity in a continuous liquid phase was defined as a standard deviation of velocity fluctuation, and the two-phase multiplier profile of turbulence intensity in the channel was clarified as a ratio of the standard deviation of flow fluctuation in a bubbly countercurrent flow to that in a water single phase flow. Finally, the distribution parameter and drift velocity used in the drift flux model for bubbly countercurrent flows were calculated from the obtained velocity profiles of both phases and void fraction profile, and were compared with the correlation proposed for bubbly countercurrent flows. KEYWORDS: two-phase flow, measurement system, multi-dimensional flow, ultrasonic velocity profile monitor, video data processing unit, bubbly countercurrent flow, velocity profile, void fraction, two-phase multiplier, turbulence intensity, probability density function, drift flux model, comparative evaluations

I. INTRODUCTION In the study of two-phase flow, the knowledge of bubbly and liquid velocities is required for better understandingof transport phenomena in two-phase flow systems. The void fraction is the most important parameter for engineering use, e.g. for the design of nuclear reactors, steam boilers, evaporating equipment, refrigeratingequipment, etc. Recently, many concepts of future light water reactors (LWRs), in which passive and simplified safety functions are positively introduced into their safety features, have been proposed such as the AP600design(1)and the SBWR design(2) in order to reduce their construction cost, to improve their reliability and maintainability and so on. However, since passive safety features are functioned by the law of nature, the driving force induced by them is much smaller than that induced by active ones. Multi-dimensional two-phase
* Oh okayama, Meguro-ku, Tokyo 152. ** W adasaki-machi, Hyogo-ku, Kobe 652. *** Wurenlingen and Villigen, SWITZERLAND . CH-5232, Villigen,

** **Uchisaiwai-cho , Chiyoda-ku, Tokyo 100.C orresponding author, Tel. +81-3-5734-3063, Fax. +81-3-5734-2959 , E-mail: maritomi@nr.titech.ac.jp
783

flow may, therefore, appear after their safety features are activated. Consequently, it is necessary with regard to passive safety features to be able to simulate multidimensional characteristics even for the two-phase flow which can be regarded as one dimensional flow for active ones. The two-phase flowshows essentially multi-dimensional characteristics even in a simple channel. The safety analysis codes such as the TRAC(3) and RELAP5(4) treat the flow basically as one dimensional flow and introduce multi-dimensional convection effects in a macroscopic way due to a lack of a fundamental data base for establishing the model of multi-dimensional two-phase flow dynamics. Therefore, it is one of the important problems to establish analytical methods of multi-dimensional twophase flow to verify analytically the effectiveness of passive safety features. It is also one of the most important subjects in the research of two-phase flow dynamics to clarify its multi-dimensional flow characteristics. Measurement methods of two-phase flow characteristics were reviewed in our previous paper(5). The authors have developed a new measurement system composed of an Ultrasonic Velocity Profile Monitor (UVP) and a Video Data Processing Unit (VDP)(5), which can measure simultaneously the multi-dimensional flow charac-

784

M. ARITOMI

et al .

teristics of bubbly flows such as velocity profiles of both gas and liquid phases and a void fraction profile in a channel, an average bubble diameter and an average void fraction. In this work, the proposed measurement system was applied to fully developed bubbly countercurrent flows in a vertical rectangular channel in order to verify its capability and to understand multi-dimensional flow characteristics in bubbly countercurrent flows. At first, both bubble and water velocity profiles and void fraction profiles in the channel were investigated statistically under various conditions of both gas and liquid phase flow rates. In addition, turbulence intensity in a continuous liquid phase was defined as a standard deviation of velocity fluctuation, and the two-phase multiplier profile of turbulence intensity in the channel was clarified as a ratio of the standard deviation of flow fluctuation in a bubbly countercurrent flow to that in a water single phase flow. Finally, concerning the drift flux model, the distribution parameter and the drift velocity were calculated directly from these measured profiles. II. EXPERIMENTAL PROCESSING APPARATUS METHOD AND DATA

Fig.

Test section

1. Experimental Apparatus Figure 1 shows a schematic diagram of an experimental apparatus. Air and water were used as working fluids. The experimental apparatus was composed of a water circulation system, an air supply system, a test section and a measurement system. The test section was a vertical rectangular channel of 100mm in width, 10mm in deep and 500mm in height made of Plexiglas as shown in Fig. 2. The measurement system consisted of the UVP, the VDP and a personal computer to record and treat data. Water was fed into the upper tank and flowed downward in the test section. The water level in an upper tank was kept constant with an overflow nozzle which was connected to a lower feedwater tank. The flow rate was measured by an orifice flowmeter and regulated by a flow control valve which was installed at the downstream

Fig.

A schematic

diagram

of experimental

apparatus

end of the test section. Adopting this flow control system, the water flow rate could be kept constant for hours. Micro particles of Nylon powder were suspended in water to reflect ultrasonic pulses. Water temperature was kept constant by a subcooler. The air supply system consisted of a compressor and a pressure regulation valve. Bubbles were injected from three needles located near the bottom of the channel. The air flow rate was measured by a float flowmeterand regulated by another flow control valve. As a result, the air flow rate could be kept constant for hours. Pressure transducers and thermocouples were installed at several points in the loop to monitor the flow condition. A personal computer acquired the readings from these sensors for an on-line control of the experimental condition. An ultrasonic transducer was installed on the outside surface of the front wall of the channel with a contact angle (t) of 45d and a gap between the transducer and the wall was filled with a jelly to prevent a reflectionof ultrasonic pulses on the wall surface as shown in Fig. 2. After both air and water flow rates were set up at the desired values, 9,216 (1,024x9) velocity profiles along a measured line were measured by the UVP under one experimental condition to treat them statistically. It takes about 30 minutes to get them. The hydrostatic head was simultaneously measured as a pressure drop betweenthe pressure taps installed on the side wall using a differential pressure transducer to get an averaged void fraction. A video camera equipment consists of an 8mm video camera, a light source and a translucent sheet to unify the luminance brightness. The speed of 60 frames per second can be obtained. After the process was videotaped, the video digital data were recorded in a personal computer through an image converter. The picture elements are 640x240 dots, the color is monotone, and the brightness resolution is 1/256. Since the detailed information about the data-handling method of the VDP was reported in our previous paper(5), its outline is described in this paper. To modify uneven brightness in the whole picture, the picture in a liquid single phase flow was taken as a standard one, and the differential picture of a bubbly flow was made by subtracting a brightness value obtained from the standard one. The data of the differential picture
JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

Measurement

System

of Bubbly

Flow

Using

Ultrasonic

Velocity

Profile

Monitor

785

wereconverted into two-valued variables, 0 and 1, by a threshold value: 0 means a liquid phase and 1 does the interface.Variables at every point in the inside surrounded bythe interface were replaced with 1. For a narrow channeland a low void fraction, assuming that bubbles were notoverlappedin the measured line and that the horizontal cross section of each bubble was a circle, the number ofbubbles in a control volume was counted, each bubble volumewas calculated and the average void fraction in the channel was obtained by summing them. Assuming that a bubble was spherical, an average bubble diameter wasobtained from the number of bubbles and the sum of the bubble volume in the control volume. In this work, 5 measuring points were adopted and the results showed that the velocity profiles of the bubbly countercurrent flow in central 3 measuring points were almost same, the velocities near the wall 2 measuring pointswere a little lower than those in central 3 measuringpoints. Therefore, the velocity profiles of 5 measuringpoints were measured to calculate local void fraction profiles. The experimental conditions are tabulated in Table 1. 2. Data Processing Method Sincethe detailed information of the proposed measurement system was also reported in our previous work(5), its measurement principle is not described and data processing method will be further illustrated in this paper. As the sound speed of the longitudinal wave is the most fundamental parameter for this method, it is not possible to treat a two-phase medium as a homogeneous singlephase medium. Figure 3 shows typical patterns of velocityprofiles, Fig. 3(a) displays a typical result of

a measuring ultrasonic beam without bubbles. Fig. 3(b) shows a typical result of a measuring ultrasonic beam with single bubble and Fig. 3(c) indicates a typical result of a measuring ultrasonic beam with multiple reflection. Since the echo of the ultrasonic pulse from the particle suspended in water is very weak intensity, multiple reflection does not appear among the particles. This fact is confirmed from measurement of velocity profiles in a water single phase flow. However, the echo from the bubble surface is not weak. Therefore, if the pulse intensity is strong, multiple reflection occurs between the bubble surface and the particles. At first, we tried to optimize the pulse intensity as the lower value whose the echo from the particle can be detected. For this optimized pulse intensity, multiple reflection is not induced between the bubble surface and the particles. This optimization was confirmed from measurement in bubbly countercurrent flows as shown in Fig. 3(b). As a rule, because a sound wave experiences multiple reflection among bubbles and its path returning to the transducer cannot be straight as shown in Fig. 4. It is however possible to obtain velocity profiles of liquid phase until the position of the nearest bubble from the transducer. Therefore, the authors attempted to derive information from each individual profiles by analyzing their shapes. Since the velocity information is derived from Doppler shift frequency, no data must be available for the wall which is at rest. The diameter of an ultrasonic pulse beam is 5mm and an UVP transducer must be inclined to the wall in order to measure velocities in the flow direction. Figure 5 shows a typical result of frequency of data existence in different setting angles of transducer to the wall. It is difficult to determine the wall position

Table

Experimental

conditions

Fig. VOL . 34, NO. 8 , AUGUST

A typical

pattern

of velocity

profiles

1997

786

M. ARITOMI

et al .

directly from the UVP data. The center of the channel is determined from measurement of a velocity profile in water laminar flows. The wall position was evaluated from the sound speed of water. It is therefore possible to identify the wall position in the profile themselves. In practice, the wall position is defined as a location where the probability of data existence is 50% in liquid single phase flow as shown in Fig. 5. Since the diameter of an ultrasonic pulse is 5mm, the measuring cross section of an ultrasonic pulse is a circular one in main flow region as shown in Fig. 6 and the echo of the ultrasonic pulse can be completely measured in the form of the circular cross section. However, the measuring cross section of the ultrasonic pulse was not a circular one near the wall region as shown in Fig. 6 and the echo of the ultrasonic pulse cannot be completely measured in the form of the circular cross section. Consequently, the measured position information near the wall region should be corrected. Figure 6 shows the measuring cross section of the ultrasonic beam in B-B position. Since the statistical average position of the measuring cross section of an ultrasonic pulse is not the center of the ultrasonic pulse near the wall region because the ultrasonic beam cannot be completely reflected in the form of the circular cross section. Assuming that the statistical average position C-C of the measuring cross section of an ultrasonic pulse divide the measuring cross section up two equal part. The statistical average position C-C is determined by SB=pR2-(bR2-R2sinbcosb), SC=gR2-R2singcosg, and SC=SB/2, (3) (1) (2)

Fig.

The

measuring

cross

section

of the ultrasonic

beam

50% of the whole ultrasonic beam one are omitted. The measurement error of the wall position is estimated as 0.1mm and the measurement error of the location +is estimated as +-0.6mm. A probability density function includes the velocity information of both phases. It is assumed that each probability density function of both phases can be expressed by a normal distribution as follows:

where SB is the measuring cross section area of an ultrasonic beam in B-B position, R is the radius of ultrasonic beam, and SC, b and g are shown in Fig. 6. From Eqs. (1), (2) and (3), b and g are calculated numerically by the Newton's method. Thus, an accurate statistical average position C-C can be obtained. The data at the position where measuring cross section area is less than

(4)
The given probability by density function of mixture velocity is

Fig.

Definition

of the

wall position

JOURNAL

OF

NUCLEAR

SCIENCE

AND

TECHNOLOGY

Measurement

System

of Bubbly:

Flow

Using

Ultrasonic

Velocity

Profile

Monitor

787

(5) where uG, uL, sG and sL are average velocities and standarddeviations of both phases respectively, e is the probabilityof bubble existence. These five variables, uG, uL, G, sL,and e, are calculated numerically s by the least squaresmethod. An example of the probability density function obtainedfrom the UVP data is representatively shown in Fig. 7. It is difficult to derive the genuine information underhigh void fraction conditions because the multiple reflectionof ultrasonic pulse is induced by bubbles. Moreover, very little information on bubble velocities can be obtained at very low void fractions. To solve these problems,several data processing programs were developedin this work(5). These programs are described below: Sincebubbly countercurrent flows are dealt with in this work, positive velocity data means bubble upflow velocity and negative velocity data does water downward one. The PROC1 code is a program which selects positivevelocity data before the position where the maximumbubble velocity appears and cuts off them behind the position in order to eliminate wrong data induced by a multiple reflection under conditions of high void fractions. Treating the data shown in Fig. 7 with this program,the results are given in Fig. 8(a). The PROC2 codeis another program to pick out only the maximum bubblevelocity in one profile as bubble velocity data and cuts off other positive velocity data. Figure 8(b) illustrates the results treated with this program. The PROC3 codeis another program to select only profiles including bubble velocities and is effective under conditions of very lowvoid fractions. The results obtained by this program is shownin Fig. 8(c). It is seen from Figs. 7 and 8(a), (b) and (c) that the mixture velocity of both phases in the probabilitydensity function does not change even if the originaldata are treated with the PROC1, PROC2 and PROC3codes. Therefore, these programs were used only

Fig.

A typical density

experimental

result

of

the

probability

function

to get the average velocity of gas phase and the standard deviation of gas phase in the probability density function of bubble velocity. The FREQUENCY code is a program which analyzes pulse height of measured velocities to give a velocity probability distribution at each point from the results for a measured profile. Since zero velocity cannot be distinguished from the data when the reflection wave is not received, the probability of the velocity number of 0 is substituted by averaging the values for the velocity numbers of -1 and 1. The SUM code sums every probability distribution calculated by the FREQUENCY code through one experimental condition. The VELOCITY code picks up a probability distribution at one point from the data obtained by the SUM code and converts the location and velocity numbers into the real position and velocity as shown in Fig. 7. The SEPARATE code is used to calculate those variables in the probability density function. Those variables are calculated numerically and iteratively by the least squares method(5).

Fig. VOL . 34, NO. 8, AUGUST

8 1997

The

probability

density

function

treated

by the data

processing

codes

788

M. ARITOMI

et al .

III.

RESULTS

AND DISCUSSION

1. Velocity Profiles of Both Phases Velocity profiles of both phases in the channel were measured with the UVP. The experimental results are shown in Fig. 9. Since it is necessary to correct the positions near the wall with significant accuracy due to an ultrasonic beam diameter of 5mm, they are corrected in the figure. Water velocities become higher toward the center of the channel from the wall in the same tendency as a water single phase flow. In contrast with this, bubble velocities are higher near the wall than those in the core. Relative velocity, which means a difference between bubble and water velocities, are shown in Fig. 10. It can be seen from Fig. 9 that the flow characteristics of bubbly countercurrent flows is strongly dependent on the water velocity which is a continuous phase and that a bubble rising velocity is induced by the difference between the buoyancy and interfacial drag force. As a result, in fully developed bubbly countercurrent flow, it can be regarded statistically that bubble diameters are almost same so that the relative velocities are nearly equal in the whole channel. Since air flow rates are much lower than water ones under the present conditions, the velocity profiles

of both phases are scarcely varied even if an air flow rate increases. With an increase in a water flow rate, water velocities become higher but their profiles are hardly influenced. In addition, the relative velocities are almost constant in the whole channel and are scarcely varied with changes in both water and air flow rates. 2. Void Fraction Profiles The hydrostatic head is obtained from the measured differential pressure and average void fraction is calculated by

(DP/DZ)Head=rG<a>g+rL(1-<a))g,

(6)

because friction loss is negligibly small due to low water flow rates. Since it was clarified in our previous work(5) that the threshold brightness, which identifies the bubble surface, influences evaluation of the average bubble diameter and void fraction even though it does not affect the number of bubbles, the threshold brightness was calibrated. A comparison of average void fractions obtained by the hydrostatic head and by the VDP is shown in Fig. 11. The void fractions obtained by the VDP agree well with those obtained by the hydrostatic head within 5% error. +A bubble diameter is calculated from its volume by assuming that it is spherical. The relationship between average bubble diameters and average void fractions is shown in Fig. 12. With an increase in a water flow rate, an impact pressure at bubble generation nozzles becomes larger and a bubble rising velocity decreases as shown in Fig. 9, so that the average bubble diameter becomes larger. In addition, as an air flow rate increases, the average bubble diameter increases because the air pressure in the bubble generation needles is enlarged. Figure 13 shows the experimental results of void fraction profile in reference to air flow rates and water flow rates, respectively. It can be seen from the figure that void fraction profile is almost flat in bubbly countercurrent flows except for those near the wall. Since air flow rates are much lower than water ones under the

Fig.

Velocity

profiles

of both

phases

Fig. Fig. 10 Relative velocity profiles

11

A comparison by hydrostatic

of average head with

void those

fractions

measured

by the VDP TECHNOLOGY

JOURNAL

OF

NUCLEAR

SCIENCE

AND

Me asurement

System

of Bubbly

Flow

Using

Ultrasonic

Velocity

Profile

Monitor

789

Fig.

12

Average

bubble

diameter

Fig.

14

Typical standard fluctuation in both

deviation phases

profiles

of

velocity

Fig.

13

Void

fraction

profiles

presentexperimental conditions, water velocity profiles are scarcelyvaried even with a change in air flow rates. Moreover, bubble velocity is dependent on the water velocityprofile as shown in Fig. 9. The void fraction is, therefore,enlarged with an increase in air flow rates. As the water flow rate increases, the bubble rising velocity is decreased,so that void fraction becomes larger. 3. Turbulence Intensity Profile In general, turbulence intensity in a bubbly flow is larger than that in a liquid single phase flow because bubbles agitate the flow. In this work, turbulence intensity is defined as a standard deviation of water velocity fluctuationin a continuous phase, sL. The standard deviationprofile in the channel can be calculated from the equation of the probability density function of mixture velocity defined by Eqs . (4) and (5). Typical results of a watersinglephase flow and a bubbly countercurrent flow areshownin Fig. 14 , respectively. In a water downward flow,the turbulence intensity has the maximum value near the wall and becomes lower with going toward the centerof the channel because the gradient of longitudinal velocities is higher near the wall. On the other hand, in a bubblycountercurrent flow, the turbulence intensity beVOL. 34 , NO. 8, AUGUST 1997

comes higher with going toward the center of the channel and has the maximum value in the center of the channel. Fluctuation of bubble upflow in the core is larger than that near the wall because the restriction of the boundary is weakened. This fact indicates that bubbles agitate the flow in a continuous phase. Since local velocities were measured not at a point but on the area because of an ultrasonic beam diameter of 5mm, the absolute value of the standard deviation in a water phase is not significant. Hence, the standard deviation ratio of a bubbly countercurrent flow to a water single phase flow is selected as two-phase multiplier of turbulence intensity, sLTPF/sLSPF. The results are shown in Fig. 15. The two-phase multiplier of turbulence intensity becomes larger with going toward the center of the channel. It can be seen from Fig. 15 that LTPF/sLSPFis enhanced with an increase in air or water s flow rates. 4. Distribution Parameter and Drift Velocity The drift flux model proposed by Zuber and Findlay(6) is applied widely to two-phase flow analysis codes. The following notations are introduced:

Fig.

15

Turbulent intensity countercurrent flow

multiplier

profiles

in bubbly

790

M. ARITOMI

et at.

(7) and (8) where F is a variable and A is a flow channel cross section. In the drift flux model, local drift velocity, vgj, and the distribution parameter, C0, are defined as follows: vgj=uG-j, and C0=<aj>/<a><j>, (10)
Fig. 17 Drift velocity countercurrent of the flow drift flux model in bubbly

(9)

where j is superficial velocity of two-phase mixture and defined by j=jG+jL=auG+(1-a)uL. (11)

Since it was difficult to measure velocity profiles of both phases and void fraction profile directly, in many previous works, average void fractions were measured under various conditions of <jG>and <jL>, and C0 and Vgj were determined by (12)
In this fraction given by work, profile velocity can profiles of both Local phases drift and velocity void is

eter is almost 1.0. Substituting the properties of air and water into the correlation proposed by Zuber and Findlay(6), Vgj=0.231m/s. The results shown in Fig. 17 are identical to this value. IV. CONCLUSIONS

be measured.

Vgj(y)=[1-a(y)][uG(y)-uL(y)].

(13)

Substituting experimental results of uG(y), uL(y) and (y) into Eqs. (10), (11) and (13), the distribution a parameters and the drift velocity were calculated directly from the measuredvelocity profilesof both phases and void fraction profiles,and the results are shown in Figs. 16 and 17, respectively.It can be seen from Fig. 13 that a void fraction profile is nearly flat in bubbly countercurrent flows. Consequently,the distribution param-

The measurement system composed of an Ultrasonic Velocity Profile Monitor and a Video Data Processing Unit was applied to fully developed bubbly countercurrent flows in a vertical rectangular channel and the following insights are clarified: (1) Velocity profiles of both phases, void fraction profiles and two-phase multiplier profiles of turbulence intensity in the channel, an average bubble diameter and average void fraction can be measured simultaneously with the proposed measurement system. (2) Water downward velocities become higher with leaving the wall but bubble rising velocities decreasebecause of higher water velocities. The relative velocities between both phases are scarcely varied in the channel. (3) Void fractions in the channel are almost constant except for those near the wall. (4) For bubbly countercurrent flows, the turbulence intensity is greater than that in liquid single phase flows and increases with going toward the center of the channel from the wall. (5) Concerning the drift flux model for bubbly countercurrent flows in a vertical rectangular channel, the distribution parameter is 1.0 and the drift velocity is the same value as proposed for bubbly upflows.
[NOMENCLATURE] A: C0: F: F>=SAFdA/A: Ft=<aF>/<a>: Flow channel cross parameter section in (mm2) drift flux model

Distribution Point Average Weighted quantity < value s mean

value of gravity of (m/s2) two-phase mixture

Fig.

16

Distribution parameter bubbly countercurrent

of the flow

drift

flux

model

in

g: j:

Acceleration Superficial

velocity

JOURNAL

OF NUCLEAR

SCIENCE

AND

TECHNOLOGY

Measurement

System

of Bubbly

Flow

Using

Ultrasonic

Velocity

Profile

Monitor

791

(m/s) center of the ultrasonic beama Superficialvelocity of gas phase (m/s) : Probability of bubble existencet Superficialvelocity of liquid phase jL (m/s) : Angle of transducer to the flow direction Pressure (Pa) : Density of gas phase (kg/m3) rG Pressure drop D between the pressure taps L: Density of liquid phase r (kg/m3)sG : Standard deviation of gas phase (m/s) (Pa) R: Radius of an ultrasonic beam (mm) L: Standard deviation of liquid phase s (m/s)sLSPF SB: Measuring cross section of an ultrasonic : Standard deviation of water single phase flow (m/s) beam (mm2) F: Standard deviation of liquid phase in two-phase sLTP flow SC: Virtual measuring cross section of an (m/s) ultrasonic beam (mm2) ACKNOWLEDGMENT u: Velocityof two-phasemixture (m/s)u : Mean velocityof two-phase mixture (m/s)uG This work was performed at the Tokyo Institute of : Mean velocity of gas phase (m/s)uL Technology in collaboration with the Tokyo Electric : Mean velocity of liquid phase (m/s) Power Company and the Paul Scherrer Institut. vgj: Local drift velocity in drift flux model -REFERENCES(m/s) Vgj: Mean drift velocity in drift flux model (1) Tower, S. N., et al.: Nucl. Eng. Des., 109, 147-154 (m/s) (1988). y: Coordinate along the deep direction (mm) (2) Duncan, J. D.: Nucl. Eng. Des., 109, 73-77 (1988). Z: Position difference between the pressure D (3) Liles, D. R., et al.: NUREG/CR-0665, (1979). taps (mm) (4) Ransom, V. H., at al.: NUREG/CR-1827, (1981). (Greek symbols)a (5) Aritomi, M., at al.: J. Nucl. Sci Technol., 33, 915-923 : Localvoid fractionb (1996). : Anglebetweenthe statistical averageposition and the (6) Zuber, N., Findlay, A.: Trans. ASME, J. Heat Transcenter of the ultrasonic beamg fer, 87, 453-468 (1965). : Anglebetween the virtual average position and the jG: : P: P:

VOL. 34

, NO.

8, AUGUST

1997