Journal By I Gusti Ngurah Bagus Catrawedarma S.T.,M.Eng Poliwangi Lecturer

1. Some Methods to Know Effect of Gap Size on Counter Current Flow Limitation
in Vertical Rectangular Narrow Channel
IGN. Bagus Catrawedarma
1
, Indarto
2
, Mulya Juarsa
3
1
Faculty of Engineering, Indonesia Hindu University, Bali, Indonesia
2
Mechanical and Industrial Engineering Department, Faculty of Engineering, Gadjah Mada University, Indonesia
3
Indonesia National Nuclear Energy Agency, Tangerang, Indonesia
Email: ngurahcatra@yahoo.com
Abstract
Effect of gap size on counter current flow limitation (CCFL) in vertical rectangular channel was studied from
experimental result. It was be conducted by using two vertical plates with 1 mm, 2 mm, and 3 mm narrow gap.
Initial temperature of the plate is about 500o
C. Debit and temperature of cooling water were controlled about
0,09 l/s and saturated temperature. As the result showed that some methods can be used to know the existence
of CCFL in the vertical rectangular narrow channels. The methods that used in this work are by comparing of
vapor contraction time between without heating and with heating, by comparing the vapor and the water
superficial velocity, by rewetting time, and by comparing CHF this experiment to other correlation. For all of the
methods are concluded that the smaller gap size, the stronger CCFL effect.
Keywords: narrow gap, gap size, superficial velocity, rewetting time, CHF, CCFL
Nomenclature
Symbol Means Dimension
ℎ
∗
∗
A
c
Dh
dT/dt
m
Nu
q
Tinitial
TC
w
ΔTexcess
Flow area
Bond number
Wallis constant
Specific heat of plate
Plate thickness
Hydraulic diameter
Transient temperature
Leinhard constant
Gravitational acceleration
Heat of evaporation
Vapor superficial velocity
Liquid superficial velocity
Non dimensional superficial velocity of vapor
Non dimensional superficial velocity of liquid
Thermal conductivity of vapor
Characteristic length
Heated length
Wallis constant
Nusselt number
Water flow rate
Heat Flux
Critical Heat Flux
Initial plate temperature
Thermocouple
Width of plate
Excess temperature
Gap size
m
2
-
-
J/kg.
o
C
mm
mm
o
C/s
-
m/s
2
W/m
2
.
o
C
m/s
m/s
-
-
W/m.
o
C
m
m
-
-
L/s
W/m
2
W/m
2
o
C
-
m
o
C
mm

2. 1. INTRODUCTION
Three mile island 2 (TMI-2) nuclear accident is one of the case of CCFL effect in narrow channel. It is
formed between debris and wall reactor pressure vessel (RPV). Based on Juarsa [2] opinion that the debris has
been cooled by water in the bottom reactor and it can’t exit from RPV wall, so reactor condition can be controlled.
When the debris move to bottom, then a part of cooling water volume is moving to upper part and evaporation
Plate density
Vapor density
Liquid density
Surface tension
kg/m
3
kg/m
3
kg/m
3
N/m

3. process is occurred. The vapor is moving to upper part but the cooling water is continually flowing to the bottom,
then it is occurred counter current flow between vapor and cooling water (see Figure 1). So, effect of CCFL in
narrow channel is very important to be considered during cooling process in nuclear reactor.
Figure 1. Narrow gap in the nuclear reactor [5]
Many experiments about CCFL in narrow channel have been conducted by researchers. Sudo and
Kaminaga [6] was investigating characteristic of CHF based on CCFL correlation and experimental CHF. It was
conducted by water as fluid test. The sizes of channel are 750 mm of length, 50 mm of width, 2.25 mm of gap
and 375 mm of length, 50 mm of width, 2.80 mm of gap. The experimental results are indicating that aspect ratio
has important role to CCFL and CHF characteristics. It was strongly implied that the CHF for downward flow is at
a minimum under the flooding condition in the case of large inlet water sub cooling and when the inlet downward
water mass flux is greater than that under the flooding condition in the case of small inlet water sub cooling.
Murase et.al [4] was analyzing heat transfer in narrow channel based on data from other researchers. It shows
that heat flux in narrow channel is larger than pool nucleat boiling in low superheat condition. It is caused by
restricted flow area and it reached CHF by CCFL. Zhang et al [8] was studying boiling curve in vertical annulus
narrow channel heated from one side. The initial temperatures were variated from 500°C to 800°C and the gap
sizes were changed from 0.5 mm to 7.0 mm. The results indicated that the heat transfer during cooling in the
channel was significantly limited by the CCFL. Under the same initial temperature, if the smaller gap size, the
longer rewetting time and the lower CHF. It is caused by existence of CCFL in the channel. Juarsa and
Antariksawan [2] were investigating heat transfer in vertical annulus narrow gap heated from one side. The initial
temperature is 800°C. 1.0 mm and 4.0 mm gap sizes are used. The results indicated that boiling in the 1.0 mm
gap size is so influenced by CCFL.
The previous experiments are indicating that the experiments about cooling process in the narrow gap are
one of the complex analisis because it is influenced by some variables. Therefore, this study is focusing effect of
gap size to CCFL in vertical rectangular narrow channel heated from both sides. Existence of CCFL is analyzed
by comparing the time of flowing water in the channel from starting process until it exit from the channel, by
rewetting time, by superficial velocity, and by value of CHF. The time of flowing water is compared between
testing time without heating and with heating.
2. THEORETICAL
2.1 Superficial velocity
Existence of CCFL based on superficial velocity is calculated by Wallis [7], that is:
∗ /
+ ∗ /
= (1)
where:
∗
=
. . ( − )
/
(2)
∗
=
. . ( − )
/
(3)
and:
= (4)
L= heated length of channel (m)
t = average time of water to flow through the rectangular channel during the heated condition.
The constant c and m are calculated by:
= 0.66
/
(5)
= 0.5 + 0.0015 . (6)

4. Where:
=
. . . ( − )
(7)
2.2 Heat flux correlation
Heat flux is rate of heat transfer per area. The correlation used in this work is:
= . . . (8)
and are obtained from properties of stainless steel.
2.3 Other correlations at each boiling regime
a. Bromley [1] correlation is used in film boiling regime. It was be got from pool boiling experiment. It is:
=
. . − ∆ℎ
.
,
× ∆ ,
(9)
C = 0,667 – 0,943
< 2
−
,
→ =
> 2
−
,
→ = 2
−
,
b. Laminer vapor correlation is also used in film boiling regime. It is:
= . ∆ (10)
Where Nu is relating to bilateral heated and geometry plate.
c. Leinhard [3] correlation is used to know value of CHF for pool boiling case:
= 0,149 . ℎ .
/
. . −
/
(11)
3. EXPERIMENTAL APPARATUS AND PROCEDURE
This experiment is a joint two facilities. They are Untai Uji Beta (UUB) and HeaTiNG-02 (see Figure 2).
UUB is used to set-up flow rate and temperature of cooling water before it is flowed to the channel. UUB has a
centrifugal pump for setting-up flow rate of the cooling water and circulating it, flow meter is used to know mass
flow rate of the cooling water, pre heater is used to increase the cooling water temperature, thermocouple is
used to measure the temperature, and some of valves is used to open or close the cooling water flow. HeaTiNG-
02 is the main test section. It has main and cover stainless steel plates with 8 mm and 3 mm thickness, 1100 mm
length, 50 mm width. Narrow gaps of the main and cover plates are changed from 1.0 mm, 2.0 mm and 3.0 mm.
The main and cover plates have three chromel-alumel thermocouples (Figure 3). It is used to measure of
temperature during heating and cooling process. Figure 2 show the schematic apparatus.

5. Figure 2. (A) Schematic apparatus, (B) Detail HeaTiNG-02
Figure 3. Thermocouple position on main and cover plate
WinDAQ T1000 Acquisition data is used to measure plate temperature during process with 1 data per
second. Slide regulator with the 25 kW maximum power is used to change power input during heating process
until the plate temperature is about 500°C. The power input was gradually increased in order to uniform
temperature.
The experiment is started by setting up the gap size with the range is showed in Table 1, then the plates
are heated by ceramic heater until the initial temperature is about 500°C. If it has been reached, the ceramic
1
2
3
5
6
4
1 : Channel
2 : Main plate
3 : Position of thermocouple
4 : Cover plate
5 : SS-316 window
6 : Heater
7 : Insulating Ceramic Brick
7
( A )
( B )
Tank
Condensor
To ECWS
From ECWS
PumpFlow meter
HeaTiNG-02
Preheater
T P
Untai Uji Beta

6. heater is switched off, then 0.09 L/s and about 90°C of cooling water that controlled at UUB is flowed to the
channel. If all of the temperatures are closing to 90°C, the experiment is stopped.
Table 1. Experimental variable
4. RESULTS AND DISCUSSIONS
4.1 Time Testing
Comparison of time testing without heating and with heating can be used as a indicator of existence of
CCFL in narrow channel (see Table 2). Time testing without heating is calculated when the plates are not heated.
It is started from cooling water flowed to the channel until it is exit from that. Time testing with heating is
calculated when the plates are heated until initial temperature is about 500°C.
Based on the comparisons of time testing without heating and with heating are known that for all of the
gap size have been existed CCFL in the channel. It is also analyzed that the longer the gap size, the faster the
cooling water exit from the channel. It is indicating that the smaller the gap size, the longer time of vapor
contraction in the channel, so the longer the time is needed by water to contact the surface plate, therefor the
stronger effect of CCFL in the channel. CCFL is the phenomena counter current flow between water and vapor
in vertical channel.
Table 2. Time testing
4.2 Superficial Velocity
Figure 4 is representing of vapor and water non dimensional superficial velocity. It is calculated by
equation (1) to (7) before it is plotted on a graph. The CCFL is occurred in the channel, if superficial velocity of
vapor is higher than the superficial velocity of water. It can be showed from Figure 4 that all of the gap size is
affected by CCFL and the 1 mm gap size has the strongest CCFL effect. It is caused by value of Jg
*
is higher
than Jl
*
. Figure 4 is also known that the larger the gap size, the smaller the CCFL effect.
Figure 4. Non dimensional superficial velocity
4.3 Transient temperature
0,020 0,025 0,030 0,035 0,040
1,4
1,5
1,6
1,7
1,8
1,9
2,0
mm
mm
mm
(j
*
g)
1/2
+ 33,48 (j
*
l )
1/2
= 2,71
(j
*
l )
1/2
(j
*
g)
1/2
Variable
Value
I II III
Gap size (mm) 1.0 2.0 3.0
Initial temperature (
o
C) 500 500 500
Cooling water mass flow rate (L/s) 0.09 0.09 0.09
Cooling water temperature (
o
C) 90.0 90.0 90.0
Tinitial
(
o
C)
Gap size ()
(mm)
Time testing
without heating
(minute)
Time testing with heating
Time of starting vapor
contraction
(minute)
Time of finishing
vapor contraction
(minute)
500
1,0 2,84 0 8,75
2,0 1,25 0 3,41
3,0 0,80 1,65 3,42

7. Rewetting point can be kown from transient temperature. Rewetting point is a condition when the cooling
water contact to the surface plate. It is indicated by drastically drop temperature. Figure 5 is showing that the
smaller the gap size, the longer the rewetting time. It is caused by the smaller mass flow rate of water cooling, the
longer the time of vapor contraction in the channel, so the longer the contact time of cooling water to plate
surface. Therefor, the smaller the gap size, the stronger effect of CCFL in the channel.
Figure 5. Transient temperature main plate on TC-6
4.4 Heat Flux
Boiling curve is a representation of heat flux versus ∆Texcess (see Figure 6). Heat flux is calculated from
equation (8) by the transient temperature. It is experimental result. Generally, in the film boiling regime show
that the larger the gap size, the higher the value of heat flux. Film boiling is occurred in the first process. It
keeps to continue until the cooling water contact to the surface of plate. The longer time the film boiling, the
longer the CCFL effect. It is caused by vapor blanket on the plate surface and it is flowing to the upper part
of the channel. So, more inconvenient the water to contact on the surface plate. The smaller the gap size,
the longer the boiling regime, and the stronger the effect of CCFL. It is also known from comparison of film
boiling regime this experiment to Bromley [1] correlation. The larger the gap size, the closer to the Bromley
[1] correlation. It is indicating that the longer the gap size, the closer to the pool boiling. It is caused by the
Bromley [1] experiment was conducted on pool boiling condition. So, the larger the gap size, the lower effect
of CCFL. In other hand, effect of CCFL can also be known from CHF value. The larger the gap size, the
higher the CHF value and the closer to Leinhard [3] correlation. It is indicating that the larger the gap size,
the closer to pool boiling condition because Leinhard [3] conducted the experiment in pool boiling case. It is
characterized by unflowing fluid. So, the larger the gap size, the lower the CCFL effect.
Figure 6. Boiling curve of main plate on TC-6
5. CONCLUSIONS
0 100 200 300 400 500 600 700
0
100
200
300
400
500
Rewetting point
Rewetting point
TC-6, Tinitial
= 500
o
C
 = 1 mm
 = 2 mm
 = 3 mmTemperature(
o
C)
Time (second)
Rewetting point
100
10
1
10
2
10
3
 = 1 mm
 = 2 mm
 = 3 mm
Bromley
Laminer
Leinhard
TC-6 ; Tinitial
= 500
o
C
q(kW/m
2
)
excess
(
o
C)
Film Boiling regime

8. Some methods can be used to know the existence of CCFL in the rectangular narrow channels. The
methods used in this work are by comparing of time vapor contraction between without heating and with heating,
by comparing the vapor and water superficial velocities, by rewetting time, and by comparing CHF this
experiment to other correlations. So, for the all of method is concluded that the smaller gap size, the stronger
CCFL effect.
ACKNOWLEDGEMENT
Thank you very much to Mr. Ismu Handoyo and Mr. Kiswanta for help and carrying attitude. Thanks to
Ministry of Research and Technology, Indonesia Government for Incentive Research Grant 2009. Thanks to
thermal hydraulic research group, Center of Reactor Technology and Safety Nuclear, Indonesia National Nuclear
Energy Agency, Tangerang, Indonesia for their support activity.
REFERENCES
1. Bromley, L.A., Heat Transfer in Stable Film Boiling, Chemical Engineering Program (1950), Vol.46,
pp.221.
2. Juarsa, M., dan Antariksawan, A.R., Effect of Counter Current Flow Limitation on Boiling Heat Transfer
in A Narrow Gap, Tri Dasa Mega Journal of Nuclear Reactor Technology (2007), Vol. 10 No. 1.
3. Lienhard, J.H., , A Heat Transfer Textbook, Third edition, Phlogiston press (2001).
4. Murase, M., Kohriyama, T., Kawabe, Y., Yoshida, Y and Okano, Y., Heat Transfer Models in Narrow
Gap, Proceedings of ICONE 9, Nice (France) (2001), 8 – 12 April.
5. Riyono B, Indarto, Sinta T.H, Mulya Juarsa Analisis Eksperimental Fluks Kalor pada Celah Sempit
Anulus Berdasarkan Variasi Temperatur Air Pendingin Menggunakan Bagian Uji HeaTiNG-01,
Publication text, Gadjah Mada University, 2010
6. Sudo, Y., and Kaminaga, M., A CHF Characteristic for Downward flow in A Narrow Vertical Rectangular
Channel Heated From Both Sides, International Journal Multiphase Flow, (1989) Vol. 15, No. 5, pp. 755-
766.
7. Wallis, G. B., One-dimensional Two-phase Flow, McGraw-Hill, United States of America (1969).
8. Zhang, J., Tanaka, F., Juarsa, M., Mishima, K., Calculation of Boiling Cuves during Rewetting of a Hot
Vertical Narrow Channel, The 10
th
International Topical Meeting on Nuclear Reactor Thermal Hydraulics
(2003).