N
| OF
ORNL P
1681
-
-
-
W
4 5
150
MICROCOPY RESOLUTION TEST CHART
NATIONAL QUREAU OF STANDARDS - 1963
4
.
.
.
1681
th
Conf-650946-4
MASTER

NOVI 15
PRESSURE DROP WITH FORCED-CONVECTION BOILING OF POTASSIUM
R. P. Wichner and H. W. Hoffman
Reactor Division
LEGAL NOTICE
This report was prepared as an account of Government sponsored work. Nellher the United
Suates, nor the Commission, nor any person acting on behall of the Commission:
A. Makes ray warranty or representation, expressed or implied, with respect to the accu-
racy, complotonour, or usefulness of the information containod in this report, or that the use
of any information, apparatus, method, or procons disclosed in this report may dot infringe
privately owned righto; or
B. Assumos may liabilities with rospect to the use of, or for denuages resultiag from the
use of any information, apparatua, metbod, or process dlaclosed in this report.
A, vaod in the above, "person acting on behall of the Commission" includes any on-
ployee or contractor of the Commission, or employee of such contractor, to the extent that
such employee or contractor of the Commission, or employee of such contractor prepares,
dienominates, or provides accenı to, any information pursuant to his omployment or contract
with the Commission, or his employment with such contractor.
For presentation at the Fourth High-Temperature Liquid-Metal
Heat Transfer Technology Conference to be held September 28-29
at Argonne National Laboratory, Argonne, Illinois
DO 190b.

RELEASED MOR MAUROUNCEMEST
IN NUCLLAR SCLINCE ABSTRACTS
OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee
Operated by
UNION CARBIDE CORPORATION
for the
U. 6. ATOMIC ENERGY COMMISSION
2
PRESSURE DROP WITH FORCED-CONVECTION BOILING OF POTASSIUM
R. P. Wichner and H. W. Hoffman
Oak Ridge National Laboratory
Oak Ridge, Tennessee
ABSTRACT ..
A
m
e
r
The results are given for the pressure drop with boiling potas-
sium in forced-convection flow through a vertical, circular tube.
The data, which comprise a single series of measurements (Series. D),
were obtained in a 6-ft long x 0.270-in.-ID tube having pressure taps
at l-ft spacings over the final 4 ft of the channel. The pressure-
measurement system and its deficiencies are discussed; the primary
difficulty appears to be "aging" of the diaphragms in the pressure
transmitters over long exposure times at high temperatures. The
pressure distribution along the boiler was estimated using both homo-
geneous and Lockhart-Martinelli descriptions of the flow; the calcu-
lated exit pressures compared favorably with the measured values.
*
.
*.--
-
!
PO
--
.
..
.
*
..
: ..
......------
*
.
..
.
e .
.
.
..
*
T
....
.
..
.
*
.
..
-
.
-
.
*Research sponsored by the U. S. Atomic Energy Commission under
contract with the Union Carbide Corporation.
Tir
.-
22::
INTRODUCTION
The pressure drop associated with forced-convection boiling is of signifi-
cant concern in the design of power systems utilizing this mode of cooling and
has been the subject of intensive study. Most attention has focused on water
in a variety of geometries at high system pressures. More recently, interests
have broadened to include lower pressure systems (including subatmospheric)
and other coolants (such as the alkali liquid metals).
While we have been accumulating pressure-drop data in the course of our
heat-transfer studies with boiling potassium in upward forced flow through
vertical circular tubes, we have described our results only briefly at these
conferences. The discussion which follows considers some of the difficulties
we have encountered in making these measurements and gives a preliminary analysis
of the data.
tanto menos minuto
EXPERIMENTAL APPARATUS
e antibioticiando menos terminali to pomembno ogni
ambient
con
en este
entre los dos contemporaneamente
d
The test channels utilized in our initial studies with boiling potassium
provided only limited information on the pressure drop. Thus, both the "lowa :
flux" and "high-flux" boilers were designed with pressure taps at the inlet
and exit only. As a consequence, the resulting data?,3 included losses (of
unknown magnitude) associated with flow through an inlet mixing chamber, around
exit-end thermowells, and through contractions and expansions at the exit.
An advanced design of the "high-flux" boiler, incorporating an additional
pressure tap at the midpoint, was abandoned because of fabrication difficulties;
and a test unit of intermediate-flux capability was constructed and installed.
This new boiler was designed to provide data on both the heat transfer and the
pressure drop; a schematic representation is shown in Fig. 1. The boiler was
machined from a 7-ft length of 1/2-in. double extra heavy IPS type 347 stainless
steel pipe (0.270-in. inside diameter, 0.839-in. outside diameter). As shown in
Fig. 1, the boiler was divided into six 12-in.-long sections which were heated
radiantly by individual clanshell heaters of 1 1/4-in. inside diameter. The
heated portions were separated by 1 1/2-in.-long gaps in which the wall was
1
i
Tenuer
14
12
SR
UNCLASSIFIED
ORNL-DWG 64-7305
THERMOCOUPLE
· LOCATIONS

LOCATIONS —-
0
1
:
- TYPE 347 STAINLESS STEEL PIPE,
0.839 in. OD, 0.270 in. ID
INSULATION
WWWXXX18XXX
CLAMSHELL HEATERS (TYP).
12 in. LONG, 1/4 in. ID
@
PRESSURE-TAP LOCATIONS
VS .
.........
..
.... .
*- - ........
............
-.--.-. .......
Fig. 1. Schematic of Medium-Flux Boiler Section.
................................
thinned from the original 0.285 in. to 0.075 in. Pressure taps (see Fig. 2
for coristruction detail) were located in the final four gaps and at the
boiler exit. The pressures au positions 2, 3, 5, and 6 were measured with
Taylor diaphragm-type pressure transmitters having NaK-filled lines and
strain-gage signal converters; at position 4, the pressure was determined
with a 6-in. bronze Bourdon-tube gage maintained at the calibration tempera-
ture of 160°F. Care was taken to insure that the pressure-tap lines leaving
the boiler were horizontal and liquid filled. While cooling jackets were
provided on all pressure taps, it was found in practice that the normal heat
loss was sufficient to assure the absence of vapor in the lines.
A pair of 0.008-in.-diam Pt-Pt 10% Rh thermocouples were also welded to
the tube wall in each of the interheater gaps; the leads of these thermocouples
were brought out along the pressure-transmitting lines. Fiberirax insulation
in these regions created an essentially adiabatic zone. In addition, tube-wall
temperatures were obtained at the midpoint of each heated section by Pt-Pt 10% Rh
couples placed in 0.035-in.-diam holes drilled radially into the tube wall; the
locations of the bottoms of these wells with respect to the inside tube surface
are indicated in Fig. 2. The 8-miI the rinocouple wires were threaded through
two-hole ceramic insulators (0.031-in. outside diameter) and individually re-
sistance welded to the stainless steel at the bottom of the well. The inside
surface temperature was obtained by extrapolation from the temperatures measured
at the bottom of the thermocouple wells.
The pressure-measuring system was calibrated at 1000°F over the range 4 to
.25 psia by comparison with a precision test gage. During calibration, the liquid
potassium was held in the sump tank; and the system was filled with helium. Cal-
ibration was effected both before and after each run series. In addition, the
electrical part of the system was checked periodically by detaching the strain-
gage leads and imposing a standard millivolt signal at this point.
The Taylor devices installed on this loop were found to be unstable, ex-
hibiting significant changes in calibration over a period of time. For example,
the output of one transmitter varied over a six-month period by nearly 12%.
These changes were always in a positive direction and derive possibly from
"aging" of the diaphragins due to prolonged exposure at high temperatures,
UNCLASSIFIED
ORNL-DWG 64-7306
- AIR
THERMOWELLS
PRESSURE TAP-
-THERMOCOUPLE
:
5

THERMOCOUPLE
THERMOCOUPLE
POSITION
4-7
:
WALL THICKNESS
(in.)
2 = 0.075
x= 0.1210, y = 0.041
x = 0.2005, y = 0.051
x = 0.2146, .y=0.018
x = 0.2070, y = 0.054
x = 0.2106, y = 0.033
x = 0.2040, y = 0.024
Fig. 2. Thermocouple and Pressure Tap Details of Medium-Flux : Boiler.

occasional imposition (as during, calibration) of subatmospheric pressures,
and/or corrosion in the potassium environuent. While this factor does in-
troduce solde uncertainty in the final results, the major effect has been
annoyance rather than loss of precision.
The quality at the boiler exit was determined by passing the effluent
vapor-liquid mixture through a separator and measuring the flow rates of the
resulting vapor (after condensation) and liquid streans both volumetrically
in calibrated hold tanks and dynamically by electromagnetic flow meters.
Other features of the experimental system have been described previously,2,3
.
•
EXPERIMENTAL AND CALCULATED RESULTS
distribution with boiling potassium flowing through a vertical, circular tube
are sw:intrized in Table I along with the inlet mass flow, exit quality, and
heat inpit... Results for two typical runs are shown graphically in Figs. 3
and 4. in the first of these (Ruil-D-4, Fig. 3), a low mass flow-high exit
quality condition existed; while in the second (Run D-15, Fig. 4), the situ-
ation was reversed. Both figures present measured and predicted pressures,
fluid temperatures, and calculated qualities as a function of distance from
the boiler inlet; the locations of the six boiler heuters are also indicated.
.
: The quality variation along the boiler was calculated from an energy
balance which neglected the wall friction and the static head but retained
the kinetic energy term. A trial-and-error solution was then effected in
which the magnitude of the slip ratio was systematically adjusted to generate
an exit quality agreeing as closeiy as possible with the measured value; a
zero Inlet quality was assumed. The results are shown by the open circles in
the lower portion of Figs. 3 and W; the measured exit qualities are given by
the filled circles. Significant changes in the slip ratio were found to have
only minor effects on the calculated exit quality. This result is not sur-
prising in that for the conditions of this experiment the enthalpy term
dominates the kinetic energy term in the energy balance. As observed for the
.
P
....
.....
th
.......
.! نارو 1
اره :
ا
۱۰ ب
:
۱۰
.
:
۱۰
) .:۱۲؛ ۲) ::
ا م م م
.
۱۱۰ : ن
1 ؛
2
.
54
ه ۱۰۰ :: .را
Vertical Circular Clavel.
Inlet nass Sri! :
intai isning
Plus N2.
. ( 14 ) :۱،
. (1/.) بدن
:: :: ::
% 1 به
( نقد و برم
.. .. م ه ه مه
، ازر:
۱۰۰ .. من أر... ، ب (
را - 4
22 - ج 2
: (بين ما
۱۰
:
۰۰
- تا ۳
.
. . .
.
!
- ای! مهر ۰\
در ۲۰ را!
.
!!! . . .
سس
دا
(
و
. يان
. 1
:
به
2
.5
31.8
5.17
23.3
ا.د .7
27.
2
۱۹.
.
4
5A
ا .15
۹.۹
۹ة .3
2
۱۹.۹ .
13.6 :
۱۱.۵ :
ا.
13.4
2.0
۱۹. 4
1 0.1
بد
16.2
عنا
۹
.
۹
و
:
ا .1
15.
13.4
۱۱۰۹
12.4
و
12.3
64.8
او
3ة .۱
3
.
6
و
3
.
۹
و
2
.
1 3.1
۱
.5
با ۱۱۰
23.3
2 .
23.9
.ا.23
23.4
22.5
29.0
25.۹
۹.
24.
ة .11
۱.۹
:
3
5.
3
۹. م
23.6
10
7.0
5:07.
५५.3
5.b۹
7.6.1
26.1
2 .
.وا
با
3
.
3
23.3
21.3
22.1
۹. .
2 .۹
۹. .
23.4
. .
22.1
۱۹.۹.
. .
?.?.?
۹ة .5
به ۱۰
2
!
7.
و .13
24.1
۶.۶.
::.
3 2
ا .2
.ره
.وہ
24.1
2
د.
.ر ?
د
..:.
53.5
13
5 .
5. 57
۹. 13
.
32.5
31.0
29.1
2
.
و .۹
۰۹ از
!.؟
1.11b
2.
1.6
97.2
. .
23.5
35.5
.
.
3.1
با
2
.
2
7.
5
31.01
3
.4
3
ا. (
7. 7
ک.25
.
21
.
3
|
-
دمروا ود+ به دوره ډط *
ر
:
د
ا » we
1 .
د
: +۱
۰۰! . و ا
رو -PE 2
.
15
3
.
23 رود و از +5 با
. .:5 : ا . ا ر . ج : 27151.ت رو 280 ز 51
313.ها را ۲۲۰ ز 24

1
!־ו־יורו
.'וי־ן־־־
.;
1
73;?:
31
LLIINILL.
175 I
A 2010
O ? Gremo Tast-
UO! nantainertrielle
Best in OVS
IT
1
.
.
44-..
.
.
.
...
w
.
M
FATH!
YM
Chi NIRNI URE
1
NUAL
}
7
T
!
-
.
1:
1
QUALITY
C
.
N1
E-L
.
.
1
I
1
.
w
t.
1
1
.
.
tury
1
...
.
E
::
-
-
2
12.
Tidl.
.
Qwzboiy
..
ml
1.
.
.
•
T
I
.
.
-
!
1
.
.
en ... danni down?
بیلنسيملمممممملن
o I ---- it
h
a
fram---Imat
d hapa II. nad Hronom Hamad In--.
Heater Locations

tttttttttt
PRE $56RE
TTTT
A heasuri
O Post from Tsat
C hart-mortoselli
X 11"r? วนง
--
I
L
-
•
-
+
.
11
*
i
ill
.
..
Iiiiii
1111111111
761762?7 URE
.
.
.
OUT
ILULLLLL
.
TTTTTTTTTTTT
T-
-
-
c:"
inTTTTTTT
, 11
CY
TIK
1-11
ri.
l
-
.
JUL
-
TTTTTTTTT
FI
Diliny
II
france.com. I
arjit nagamuw- TI
-- toy pa
/7
od popo...-12
. krama 1
ud herinn- VI ---*
10
:
two cases graphed, the calculated exit quality is always somewhat lower than
the measured exit quality. Since the slope of the axial quality profile is
fixed by the known heat input, the inlet quality could be obtained by extrap-
olation back from the measured value. The inlet qualities thus obtained were
15% for Run D-4 and 10% for Run D-15. These estimated values are markedly
greater than the qualities believed, on a qualitative basis,, to have existed
at the biler inlet and may reflect in part errors (on the high side) in the
measured .. jalities due to vaporization between the boiler exit and the meas-
urement etition. In any event, the pressure-drop estimations discussed below
:
were based on the calculated qualities without adjustment for the indicated
deviatio:2 from the measured exit quality.
The central portion of both Figs. 3 and 4 show the fluid temperature dis-
tribution along the boiler. The data plotted are wall temperatures measured
in the thin-walled (0.075 in.), adiabatic zones (1.5 in. long) between the
heaters. It is presumed, under these circumstances, that the wall temperature
is essentially the same as the fluid temperature.
Pressures measured at five positions along the boiler are shown by the
open triangles in the upper portion of figs. 3 and 4; as noted in Fig. 1; the
first pressure tap (PE-6) was located in the adiabatic zone between the second
and third heaters. The data show somewhat more scatter than previously ob-
served in run Series C;4 this may result from the uncertainties associated with
the pressure-measurement system as discussed above in the section on apparatus.
An arbitrary smooth curve has been drawn through the data.
The lower pressure curve in both figures gives the saturation pressure.
(open circles) as derived from the fluid temperature data. The pressure dis-
tribution thus derived shows a trend with distance along the boiler that
agrees closely with the curve through the measured values but falls signifi-
cantly and consistently below the latter. This discrepancy in magnitude results
in part from the use of the vapor pressure correlation of Lermon et al.5 in this
analysis; at a given T.+, Lemmon and his co-workers found Poot to be less than
· the value indicated by other available correlations. Thus, using the equation
developed by Ewing et al., the saturation pressure calculated at position PE-4
1
:14.
SNA
*
in Run D24 is 1.47 atm; this compares with a value of 1.39 atm calculated from
the Lermon et al. equation and 1.54 atm measured. Similarly, for Run D-15 at
the same measurement station, Pset = 2.14 atm according to Lemmon, 2.22 atm
according to Ewing, and 2.34 atm measured. The remaining difference between
the measured pressure and that calculated from Tat could derive from low tem-
perature readings due to heat loss in the "adiabatic" zone or from a systemic
error in the pressure measurements.
For purposes of comparison, the pressure drop along the boiler was also
estimated using both the homogeneous and the Lockhart-Martinelli? models for
tained by assuming the fluid to be homogeneous, while the closed circles give
corresponding values obtained by the Lockhart-Martinelli procedure with cor-
rections for changes in the kinetic and gravity heads.
The Blasius friction-factor equation, fr = 0.0396 NO.25, was used in
the homogeneous calculations, while the analogy relation, fn = 0.023 NO.20,
was used in the Lockhart-Martinelli computation. The following further com-
ments are pertinent to the estimation of the pressure drop by the Lockhart-
Martinelli procedure:
1.. Since the slip ratio was unknown, a value of unity was assumed in
the calculations of the kinetic and gravity head corrections.
2.- For convenience in the computer calculations, an empirical equation
was fitted to the Lockhart-Martinelli parameters; thus,
en ¢ out =. 1.478 - 0.5403 en Xtt + 0.05194 (en xx+)2 + 0.000698 (en x++)3 ,
where ... is the square root of the ratio of the total two-phase pressure drop,
(aP/NL)orape to the pressure drop for the liquid phase only, (ap/aL),; and
- (-2)*(**) * (0),
x bein; a mean quality in each leated section. Over the range of the experi-
mental al'a, 0.01 < X + < 1.00, this empirical equation overestimates $ ettat 1
Xtt. = 1.0), the error is 15%, and at Xit = 0.01,: ~10%.
12
.
3. Since the pressure was not measured at the boiler inlet, the starting
point for the pressure calculations was assumed to be Past, as estimated from
the wall temperature measurements. This is probably not an unreasonable as-
sumption in that the fairly long loop section immediately preceding the boiler
is essenti!lly adiabatic; and, hence, the wall temperature measured at the end
of this region should equal the fluid temperature. The data shown in Figs. 3:
and 4 suggest this conclusion.
.
.
. .
.
.
...
DISCUSSION
In view of the error discusced above in the evaluation of dett, the
Series [ pawta have not been presented in the usual graphical form of the
pressure Wrop ratio versus the Lockhart-Martinelli parameter
These
calculations will be repeated in the near future using a newer, more precise
computer procedure for evaluating ++Thus, any comparison of the present
data with results obtained with water under similar experimental conditions
must be deferred.
The final three columns of Table 1 provide a comparison of measured and
calculated exit pressures; in general, the agreement between the tabulated
values is good. However, there does not appear to be any consistent trend
in these comparisons. Thus, for Run D-4, Lockhart-Martinelli predicts an
exit pressure less than measured, while the homogeneous model yields a value
greater than measured. In Run D-8, the situation is reversed; while in Run D-15,
both calculations indicate exit pressures greater than measured. As discussed
above, these comments must be qualified somewhat in respect to the Lockhart-
Martinelli calculation. An overestimate in the pressure drop by 15% on the
average is indicated; this results in predicted exit pressure which are 3 to
5% low.
ACKNOWLEDGMENTS
ILT
Ille authors acknowledge the contributions of A. I. Krakoviak in the design
and construction of the test section, of R. D. Bundy and B. J. Sutton in obtain-
ing the series D experimental data, and of Dolores Eden for her conscientious
typing of the manuscript.
e.
*
-
-
·
-
..
. 13
REFERENCES
1. H. W. Hoffman, "Recent Experimental Results in ORNL Studies with Boiling
Potassium," pp. 334-350, Proceedings of 1963 High-Temperature Liquid-
Metal Heat Transfer Technology Meeting, vol. I, USAEC Report ORNL-3605,
Oak Ridge National Laboratory, November 1964.
.. .-
2. H. W. Hoffman and A. I. Krakoviak, "Convective Boiling with Liquid Potas-
sium," pp. 19-37, Proceedings of the 1964 Heat Transfer and Fluid Mechanics
Institute, W. H. Gredt and S. Levy, editors, Stanford University Press,
Stanford, California, 1964.
..
......
3. A. I. Krakoviak and H. W. Hoffman, "Boiling Potassium Heat Transfer:
Forced-Convection System," pp. 19-20, Studies in Heat Transfer and Fluid
Mechanics, Progress Report for Period January 1- September 30, 1963,
H. W. Hoffman and J. J. Keyes, Jr., editors, USAEC Report ORNL-TM-915,
Oak Ridge Naticnal Laboratory, October 1964.
4. A. I. Krakoviak and H. W. Hoffman, "Boiling Potassium Heat Transfer:
Forced-Convection System," pp. 7, 12-15, Studies in Heat Transfer and
Fluid Mechanics, Progress Report for Period October 1, 1963 - June 30, 1964,
H. W. Hoffman and J. J. Keyes, Jr., editors, USAEC Report ORNL-TM-1148,
Oak Ridge National Laboratory, August 1965.
A. W. Lemmon et al., "The Thermodynamic and Transport Properties of Potas-
siun," pp. 88–115, Proceedings of 1963 High-Temperature Liquid-Metal Heat
Transfer Technology Meeting, Vol. I, USAEC Report ORNL-3605, Oak Ridge
National Laboratory, November 1964.
C. T. Ewing et al., "High Temperature Properties of Sodium and Potassium,
12th Progress Report for Period 1 July to 30 September 1963," U. S. Naval
Research Laboratory Report NRL-6094, June 1964.
R. W. Lockhart and R. C. Martinelli, "Prediction of Pressure Drop, for
Isothermal Two-Phase, Two-Component Flow in Pipes," Chem. Eng. Progr.,
45: 39 (1949).
.
.
..7
.
.
-
-
-
--
--
-
.
·
·
..
TA
2.
.
12/ 1 /65
DATE FILMED
END

-
.
-
-
.
-
-.
T
.
L
.
E
yer
est
24
HU
..
1