I
.
.
-
.
?
I OF I
ORNLP
172)

*
19:31912
0125 ||L4 ILS
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS -1963
...
--•.
-
-
-
-
..
. -..
-
4.
41
25
'n
'
B
!
!
*
.
.
:

.
VMPRE RADIATOR DEVELOPMENT TESTS*
Pont_c5701234
A. P. Promo
MASTER
A. P. Fraas
Oak Ridge National laboratory
Oak Ridge, Tennessee
NOV 1
ABSTRACT
Six tapered tube direct condensing radiators have been built and operated in the course
of the system development test program at ORNL, and a number of tests have been run with
individual tubes. The bases for the design of these units and the more significant results
of the test program are summarized, and some implications relative to future design work
are discussed.
Introduction
3.4.
i
11
ART
. The boiling potassium reactor system under development at ORNL presents a number of
fluid flow stability and control problems that involve close coupling and possible inter-
action between the boiler and condenser. Because of this, the MPRE development program
has been based on the use of a direct condensing radiator with a set of proportions rea-
sonably representative of those that have appeared suitable for a space vehicle.
WS
Six radiators have been built and operated - all as elements of systems rather than
in rigs built expressly to test radiators. No effort has been made to obtain good data
2
on condensing heat transfer coefficients or surface emissivities; our attention has been
2
.
Y
...
T
.
1
.
-
2. '
directed primarily at problems of radiator design to obtain good flow distribution, flow
stability, a sound structure, and good system stability and control. Particular atten-
tion has been given to possible interaction between the flows in tubes operating in par-
allel, and to the flows in multiple banks of tubes.
In arriving at the basic radiator design to be used in the test program, a wide
variety of configuration was considered from the standpoints of free-liquid-surface con-
trol under 0-g conditions, scavenging, fluid flow distribution, part load stability and
.
-
.
.
.
2
. ,*
.
.'.
S.::.
.
.
.
.
K
...
..
4
"Research sponsored by the U. 8. Atomic Energy Commission under contract with the
Union Carbide Corporation.

RENTIBUSUID POR ABUSOLEKCEMENT
IT ACTUEAR SCIENCE ABSTRACTS
.:
19
.
?
22.9
control, differential thermal expansion, installation in the launch vehicle, the stresses
and vibration induced by launch loads, and orbital startup. Analyses disclosed that there
were too many boundary conditions to be met to permit a conventional optimization study -
it appeared better to begin with a structure that would be well-suited to withstand launch
loads and at the same time would satisfy the fluid flow requirements. The proportions were
then modified to reduce the weight to something like a minimum; this led to the reference
design for a launch package shown in Fig. 1. The efficacy of this approach 18 indicated
by the relatively low specific weight obtained for the resulting launch package - only
7 10/kw of electrical power output. This specific weight is for the complete radiator
including meteoroid armor, ducts, and the structure for launching. The background for this
selection was outlined in ref. 1 with some discussion of the problems presented by radia-
tion scettering from the radiator and its effects on the shield design, meteoroid protec-
tion, two-phase flow in tapered tubes, and the effects of load on performance. ..
-
-
::.
--
-
-
--
--
Radiator Design
The design precepts upon which the ORNL radiators have been based can be presented by
referring to the radiator in the reference design for a launch package shown in Fig. 1.
This radiator consists of parallel banks of axially finned, tapered tubes arranged on the
perimeter of a cylinder with an Alzak aluminum reflector placed behind each of the tubes
-...
---
-
- -
to reflect heat from their rear surfaces to space as shown in Fig. 2. The reflector can
-
.
.
.
.
.
be contoured so that any ray emitted from the rear of the finned tube will be reflected
out past the tips of the transverse fins to space. Aluminum with an Alzak finish appears
to be well-suited to this application since its efficiency for speculer reflection 18
about 85%. Its equilibrium temperature at design.conditions will be about 600°F, which
18 satisfactory for anodized aluminum even in air. While this tube-and-reflector arrange-
ment represents a major departure from the approaches used by most designers, it appears
to be a sourd approach and has many advantages. Perhaps the most important of these is
that it keeps both the fin weight and the amount of Area vulnerable to meteoroids to
TEGAL NOTICE
The report me prepared N AA nooount of coverunt meword works Netther the Owind |
: Maws, ter the Commission, nor any person sotilas on behalf of the Counselo! .
.. A. Makes may warranty or representation, express or impuedo mi respect to the noot
they, completenes, or wafalmous of the information contain the report, or that the we
of my Information apparatu, methods or proouse dineloned to the report may not taarten
privately owned this or
1. nummus neng labuchos with repect to the wood, or for dem somitting trom the
w of aty tudormation, apparate, method, or proomu dieslowd la dels roporto
As wood la the abova, "person noting on behalf of the Commissioni includes my w
ploys or contracter of the Commission of maplyn of such contractor, to the accent that
nuel employto or contractor of the Commission, or comployee of mot computer prepared
download, or provide moont to, any information per tend to Memployment
nodom, Ma gnorum with a combinatorium menina. Situation and in ****
vir
.
EN
.'
?
MA
V
.
lt
1
TTL
.'

L.
wa...
.
..
..
...
.....
...
.....
moderate values. By surrounding the unit with a thin aluminum windshield, which could
be in the form of the shutters indicated in Fig. 2, and foaming plastic between the wind-
shield, the radiator tubes, and the aluminum reflector, a strong, stiff, vibration-
resistant structure would be provided to resist the dynamic loads and severe vibration of
a launch. When the power plant is warmed up in orbit, the polyurethane foara will be
Lok*"
volatilized into the high vacuum surrounding the spacecraft, and the windshield can be
removed if it is a continuous sheet around the perimeter, or flaps such as those shown in
Fig. 2 can be opened.
The design calculations for the reference design radiator are summarized in Table 1.
As shown in Fig. 2, the portion of the tube vulnerable to meteoroids was made 0.10 in.
thick to reduce the probability of meteoroid penetration to 1% in 10,000 hr, while the
rear portion of the tube was made 0.030 in. thick to give a probability of about 0.1%
that an imperfection in the metal wall will lead to a leak. This consideration stems
from extensive endurance test experience at ORNL with high temperature heat exchangers 2,3
This shows that individual grains often grow to a size of about 0.010 in., and since
there may be imperfections in individual grain boundaries, it is best to make the tube
wall thickness at least twice that of the largest grains, i.e., at least 0.020 in. thick."
The thickness was increased to 0.030 in. to increase the system integrity because the
resulting weight penalty was only 10 lb. For weight estimating purposes this gave an
average tube wall thickness of 0.065 in. To reduce the fin weights, the fin thickness
wae varied so that it would be directly proportional to the square of the distance from
the outer tip. Copper was chosen as the fin material because it is ductile, its coef-
ficient of thermal ·xpansion is close to that of stainless steel, and it is easily brazed.
Ideally, beryllium fins would give a weight savings of about 60 lb, but beryllium 18
brittle, its coefficient of thermal expansion is only two-thirds that of stainless steel,
and it is difficult to braze.
Subsequent to the preparation of Fig. 1 it became evident that it would be better
to increase the number of ducts radiating to each manifolå ring to six, and this was
as sumed in preparing Table 1. The weight of the ducts and manifolds totals 280 lb, or
-
-
28% of the total radiator weight at launch. If allowance is made for the fact that these
elexents are also designed to serve as structural members to support the turbine-generator
under launching conditions, (a 10 g vertical acceleration would induce a 10,000 psi' stress
in these parts) this weight does not appear excessive. The two sets of six radial ducts
that extend from the central hub to the two manifold rings are 3.5 in. in diameter and
have a wall thickness of 0.040 in. to resist buckling under ground test conditions where
atmospheric pressure would tend to cause them to collapse. The circular manifolds also
have a base wall thickness of 0.040 in. with an additional 0.060 in. of thickness on the
outer portion to provide meteoroid protection. These circular ring manifolds teper from
a diameter of 3.5 in. at the joints with the radial ducts to a diameter of 2 in. in the
regions midway between the joints.
In laying out the development test program for the MPRE, it became evident that a
number of component test systems would be required to investigate major problem areas, :|
and that condensers would be required for these systems. It was found that in many in-
stances an adaptation of the tapered-tube elements of the radiator of Table 1 served to
give about as inexpensive a condenser as could be obtained in any other way, and that use
of a properly designed and instrumented condenser would yield invaluable information on
the performance of this type of direct condensing radiator. Since the most important
single question that arose in the course of the design studies was concerned with the
control of the free liquid surfaces in the condenser under 0-g conditions, it was de- :
cided to build some of these units for the test rigs so that they would operate with i
the tube axes nearly horizontal. Since it was important that the liquid drain from the
tubes after the test, the condensers were built as in Fig. 3 and run with the tubes.
at a small angle from the horizontal to facilitate, drainage. It is important to note
that analytical work indicated that the drag forcee acting to move liquid toward the out-':
lets of these nearly horizontal tubes would be of the order of 100 times the gravitational
forces under. Pull power flow conditions. Units were designed so that under high power
'conditions the tube taper gives a nearly constant vapor velocity down the tube from the ...
Inlet to close to the outlet.
•
It happens that, as the power to be dissipated by the radiator is reduced, its tem-
perature must of necessity drop. Ideally, the vapor pressure of the potassium in the
radiator tubes would vary in such a way that the velocity of the vapor down the tubes
would be constant irrespective or the power plant output, and the pressure would fall off
in direct proportion to the power. If no control 18 exercised over the heat dissipation
from the radiator to space, the potassium vapor pressure will fall off more rapidly than
the heat load so that, as the power le reduced, the vapor velocity in the tubes will in-
crcase to sonic velocity. Taus, while the pressure ratio between the tube inlet and out-
let 18 quite small at full power, at part-load, if no shutters are operated to reduce the
heat dissipation, this pressure ratio will increase to the point where a substantial tem-
perature drop will occur along the length of the tube as a consequence of a series of
oblique shocks in the vapor flow down the tube.
The effects of load on the axial temperature distribution in the tubes of a direct
condensing radiator were investigated both analytically and experimentally with the heat
sink temperature as 'n parameter. The radiator temperature varied, of course, with the
heat sink temperature, but, for the radiator proportions used, the axial temperature
varied less than 30°F 1rrespective of the heat sink temperature over the load range from
100% power down to about 60% power. Some typical results of the analytical work are shown
in Fig. 4 for loads of 50%, 30%, and 10% power. Note the good agreement in Fig. 5 between
the analytical estimates and a set of experimental data for the same set of conditions.
It is essential that the pressure at the inlet to a pump be appreciably above the
vapor pressure. In ground tests It is possible to design the piping so that there will
be sufficient static head to suppress cavitation at the inlet to-the condenser scavenging
pump no matter how low the pressure at the condenser outlet. Under 0-g conditions, how-
ever, cavitation suppression head may be obtained only by subcooling the condensate.
Consequently, some means of reducing the heat losses from the condenser must be employed
at reduced power so that condensing will occur at a temperature and pressure sufficiently
high to make it possible to obtain the required cavitatiöö" suppression head by subcooling...
.
-
.
-
-
--
-
---
IL
· L'igure 4 shows that increasing the heat sink temperature is a good way to increase
the condenser outlet temperature, hence the shutter arrangement of Fig. 2 affords a prom-
ising approach. Some notion of the effectiveness of shutters can be obtained by consid-
ering some envelope cases. For example, if shutters which are black on both sides are in
a completely closed position, ideally they will cut the heat dissipation rate for a given
tube temperature in half. If the shutters are black on the inside but have polished gola
surfaces on the outside with an emissivity of 0.02, the heat dissipation at a given tem-
perature would be reduced to only 2% with the shuttere in a closed position. The shutters
will be particularly effective in reducing heat losses if their surfaces are specularly
reflective and they are mounted close to the tubes as in Fig. 2 so that most of the heat
rays emitted from the radiator tubes will be reflected back to the tubes after undergoing
only one or a few reflections.
In analyzing the heat losses as a function of shutter ang].e for the arrangement of
Fig. 2, it was found that the view factor associated with specularly reflecting surfaces
on the inside of the shutter led to a rapid change in heat dissipation for relatively small
changes in shutter angle in the vicinity of angles of 45.to 60 deg. This effect 18
obviously undesirable because it gives a highly non-linear relation between the heat dis-
dipation rate and shutter angle, a poor characteristic from the control standpoint. .A
single full-span shutter door was then considered, and this was found to give' very much
more desirable characteristics as can be seen in Fig. 6. Thus it appears that the single
full-span shutter door is definitely superior to the two half-span doors, and hence the
full-span door arrangement was chosen for further design purposes. Analyses indicate that
the radiator temperature need not be controlled within close limits; the control device
simply needs to reduce the effective emissivity and/or view factor sufficiently to keep
the condenser heat dissipation rate within a factor of two to five of the ideal value.
The design of the shutter operating mechanism presents a number of problems. Differ. :
ential thermal expansion between the shutter blades, the reflector, and the tubes must be
accommodated with a light weight installation without seriously interfering with alignment
..of the hinge points. One promising arrangement employs a flat sheet with a stiffening
: ::6
i
.:
..
-
.
*.
T
magos
-
-
."
F
-
.
PL
.
c
**
sp. mom
bead along its outer edge and its inner edge pivoted on a wire stretched between brackets
on the vapor and condensate manifolds. This would be actuated by bimetallic thermostat
elements' mounted at either end.
. One problem posed by this method of controlling the heat dissipation from the radia-
tor stems from the effects of thermal convection when ground testing a unit in air. The
initial test consisted of an electrically heated tube 4 ft long mounted vertically with a
reflector and shutter.
The stack effect with the shutter closed was so large that the
shutter was much less effective than expected. These tests are being repeated with trans-
. verse balfles at the ends designed to prevent vertical air flow over the tube.
Test Experience
It happens that steam at about 100°F 18 an excellent stand-in for potassium vapor at
about 1000°F. in that the densities, viscosities, and heats of vaporization of the two fluids
are about the same. Thus to reduce costs and to permit visual observation of the flow
through glass, and plastic parts, the initial system tests were carried out with steam.
One example of a unit designed for tests with steam is the flat slab condenser of Fig. 7.
This was designed to permit observation of the movement of the condensate film on the cooled
walls, and has eight parallel tapered passages for the steam. It was fabricated by sand-
wiching. hollow, rectangular copper bus bars between two sheets of 1/2-in.-thick Plex.iglas.
Other units built for tests with steam have included a bank of 12 tubes 7 ft long that
tapered from a 0.60-in.-ID to 0.20-in.-ID.' Using tubes of the same size, a 144-tube bank
similar to that of Fig. 3 was also built and tested in a steam system. Both of these units
were water-cooled, hence fins were not required and were not used. The pressure drop at.
full power output was so low (a few millimeters of mercury) in the tests with steam that
it. was not possible to get an accurate check on the calculated pressure drop because of
small' uncertainties associated with the pressure tap locations. However, within the
limits of experimental error the test data were consistent with the calculated values.
:: The first tests with potassium were run with the radiator of Fig. 8 which consisted
of a single bank of 12 tubes. Because of a fabrication error, the feed pump discharge
til.
-.,
.
denser and, as a consequence, the tubes vended to load up with liquid, giving an abrupt
break in the axial temperature distribution at the liquid-vapor interface close to the out-
let. When this condition was corrected, the temperature distribution in the condenser was
quite uniform at the higher temperatures. At condenser temperatures below approximately
800°F, however, the temperature pattern shown in Fig. 9 was obtained. Note that near the
outlet the outer tubes ran at temperatures almost 400°F below the inner tubes in the bank.
This condition was undesirable because the consequent differential thermal expansion
forced thermal distortion (see Fig. 10). At first the poor temperature distribution was
charged to the greater temperature losses from the outer tubes but, if this had been the
case, it is hard to see why the effect should have extended so far in from the outer
boundaries. On further examination it was decided that the effects stemmed from sonic
velocities in the vapor manifold. The vapor pipe feeding the center of the tube bank
gave higher pressures in the center tubes and hence higher temperatures. Shock losses in
the vapor manifold led to much lower pressures – and hence temperatures -- in the tubes
farther from the center feed point. This condition was not encountered with steam be-
cause the cooling water supply did not permit operation at a 10: enough temperature to
give sonic velocities in the vapor manifold. As a consequence of this experience, all
subsequent radiators have been designed so that choking will occur at the inlet to the
tubes rather than in the vapor manifold so that the flow distribution to the various
tubes will be uniform. Thus, when shock lusses do occur, they take place within the tubes
and lead to an axial pressure and temperature gredient that is essentially the same for
all of the tubes in the radiator.
The potassium vapor direct condensing radiator of Fig. 3 was designed with the
proportions indicated in Table 2 to dissipate 500 kw of heat at 1000°F. This has been
operated for over 2000 hr in a potassium system. Figure 11 shows this condenser after
installation in the heat sink, which consisted of a double-walled box designed so that
about 90% of the heat emitted from the radiator would pass by thermal radiation to the
inner wall of the box, while the remaining 10% would be dissipated by thermal convection.
-
41
4
w
momenten er de beste handi batean alder than one thing
comes the
The radiator was scavenged by two jet pumps, one serving each side. Each jet pump was
connected to six banks of tubes (12 tubes per bank) by six lines of equal length to
equalize the radiant heat losses and thus the amount of subcooling between the condensate
manifolds and the jet pumps. The principal features of the installation can be discerned
in Figs. 3 and 11. These jet pumps require a cavitation suppression head of about 0.2 1n.
of potassium at 10% load, hence the radiator would not scavenge properly if the outlet
temperature were allowed to drop below about 560°F under 0-g conditions at 10% load.
Perhaps the most significant data obtained in the course of the test program are a
series of infrared photographs taken of the condenser with no illumination other than the
infrared light emitted from the radiator surfaces. Figures 12 and 13 were taken looking
up from below at the essentially horizontal radiator (see Fig. 11). These pictures, made
while the feed pump was scavenging the radiator çroperly, indicate that uniform scaveng-
ing of the radiator was occurring, and that the radiator temperature distribution was
surprisingly uniform. An indication of the sensitivity of this technique is given by the
fact that the tenperature differential between the roots and tips of the Pins was only
25°F. Thermocouples in the field provided base points for evaluating the temperature from
the local density of the image. The radiator surfaces had not been treated to give a high
emissivity, hence light from the higher temperature vapor manifold at the center was
reflected from local areas on the fins (which had warped and wrinkled somewhat during the
brazing operation).
After examining Figs. 12 and 13 with their apparently remarkably uniform temperature
distribution, it was decided that it would be well to investigate the effect on the in-
frared pictures of loading up the radiator with a substantial amount of liquid. This was
done by partially closing a valve in the condensate return line from the condenser to the
feed pump, and the photos of Figs. 14 and 15 were taken. The system was stabilized with
- the liquid level in the expension tank at the boiler reduced by an amount consistent with
- the quantity of liquid that Figs. 14 and 15 indicate had collected in the radiator tubes.
Note the sharp drop in temperature near what must be the liquid-vapor interface in the
in the radiator tubes. Note, too, the small scallop 10 the liquid interface positions
....
.
..
..
...!
across the tube banks, with less liquid holdup in the tubes near the center of any given
114
bank of 12 tubes. Refinements in exposure time and technique should permit detection of
temperature differences less than 10°F. These pictures indicate that infrared photos will
be an extremely useful tool for determining the temperature distribution in the condenser
and the location of the liquid-vapor interface between the condenser and the condenser
scavenging jet pump. °
Several observations can be made from the infrared photographs and the thermocouple
and pressure gage data obtained. The operating pressure in the radiator as estimated from
temperature measurements was about 0.15 psia. The pressure drop, as indicated by a potas- ,
sium manometer, ran about 0.1 psi at ful). load – too little to measure accurately. The
thermocouple data and the infrared pictures indicate that vapor bubbles were leaving the
radiator tubes, passing through the condensate manifold into the outlet line, and perhaps -
in some cases - reaching the jet pumps. The unit was designed so that it should scavenge
properly with no help from gravitational forces, and the test data indicate that this
objective has been achieved when the scavenge pump has adequate capacity and the radiator
outlet temperature is kept above some minimum value that depends on the cavitation suppres-
sion head requirements of the scavenging pump. The 144 tubes have operated nicely in.
parallel with no sign of flow instabilities and no maldistribution in the flow sufficient
to cause any apparent warping of the tubes in 2000 hr of operation under a wide variety
of conditions. The latter indicates that proportioning the vapor manifold so that sonic
velocities will be reached in the tube inlets before they will be reached in the manifold
18 an effective solution to the problem cited earlier in connection with Figs. 9 and 10.
The next step in vur program will be to run a larger system with a full-scale version
of the radiator of Fig. 1 and Table 1. The appearance of the test radiator in the course
of fabrication is shown iø Fig. 16. To reduce fabrication costs the tubes for this test
unit have been made with a uniform thickness of 0.050 in., and the fins have been made of
stainless steel cled copper sheet with a uniform thickness of 0.040 in. of which 0.030 in.
lies in the copper core. The surfaces will be treated with a high emissivity coating.
-
.
.
2
.
.
? .
.
.
.
.
Another cost-saving measure was the increase in the wall thickness of the ducts and mani-
folds to about 1./8 in. to facilitate fabrication in our shops.
Table 1. Performance, Dimensional, and Weight Data for the
Radiator of the Reference Design for a Launch Package
.
:
Item
Value
860
2.93 x 106
8560: .
0.92
0.86
0.87
435
.
14.0
7.0
1500
0.0035
887
Power to be dissipated, kw
Power to be dissipated, Btu/hr
Ideal dissipation at 1500°F to 500°F sink,
Btu/hr.ft2
Emissivity of treated surface
Fin efficiency
Reflector efficiency
Area required, fta
Height of cylinder for 10-ft diameter vehicle,
ft
Tube length, ft .
Vapor temperature, 'R
Vapor density, 10/ft3
Latent heat of vaporization, Btu/10
Vapor quality, %
Potassium flow rate, 1b/sec
Vapor volume flow rate, ft3/sec
Vapor velocity at tube inlet, ft/sec
Vapor flow passage area at tube inlets, ft
Number of tubes
Tube inlet flow passage area 'per tube, fts. .
Tube inlet ID, in.
Liquid density at tube outlet, lb/ft3
Tube outlet liquid flow rate, ft3/sec -
Tube outlet ID, in..
Tube outlet area (per tube), ft? .
Tube outlet area (total), fta
Tube outlet velocity, ft/sec (for liquia).
Fin span, in.
91.7
1.0
262
400
0.655
96
0.0068
1.1
44.0
0.0228
0.30
0.0049
0.060
0:38
3.2
Table 1. (Cont.)
Item
· Value
!
t = 6(x/w)2
0.46
20.0
1:2
0.092
70
.
290
120
100
Fin cross-section shape
0.408 w Vh/kb for 90% fin efficiency
Equivalent heat-transfer coefficient,
Btu/hr• ft2.°F
Mean fin height, in.
Fin root thickness, in.
Vulnerable surface area (total), fta
Weight of copper fins, lb
Weight of 0.065-in.-average wall thickness
tapered stainless steel tubes, lb (in-
cluding 70 lb armor)
Weight of 0.010-in.-thick aluminum
reflector, lb
Weight of 3.5 in. diam radial vapor ducts
15.2 ft long, 0.040 in. thick, lb
| Weight of ring manifolds, lb
Weight of armor for ring manifolds, lb
Weight of inlet duct and hub, lb
i Weight of 0.010-in.-thick aluminum shutters
with operators, lb
'Weight of 1.5-1b/ft3 polyurethene foam, lb
Weight of aluminum meteoroid bumper at top
end, lb
Total launch weight, lb
Net weight after vaporizing polyurethane, lb
95
.
50
100
100
10
1000
900
Table 2. Design Data for the 144- Tube Radiator for the Intermediate
Potassium System*
Design power output
Operating temperature, °F
Black-body heat flux, Btu/hr.ft2
Fin material
W
500 kw
1040
8600
ss-clad
copper,
0.75
0.010
(0.006 in.
Cu).
Etai ssivity
Fin thickness, in. (total)
.
700
0.60
76
Heat transfer coefficient for thermal convection,
Btu/hr.ft2.°F
Heat flux for thermal convectio., Btu/hr.ft2.°F
Fin height, in.
Fin efficiency, % .
Tube OD (mean), in.
Tube OD at manifold, 10.
Tube OD at exit, in.
Tube wall thickness, in.
Fraction of surface in fins .
Average fin heat flux, Btu/hr.
Average heat flux, Btu/hr.ft2
Surface area, ft2
Average surface, ft/ft of tube length
Total length of tubing, ft i
Number of tubes
Tube length, ft
Manifold d:lameter at base, in.
Manifold diameter at tip, in.
*The radiator is shown in Figs. 3 and 11.
0.57
0.80
0.30
0.050
0.636
4700
5500
298
0.295
1008
144
|
TAR
-
.
BV

.
WY
L
u
?
.
ho
tu
.
1
L
.
.
.
.
k
.
.
.
"
•
IN
'
'
:
.
.
UN -
.
IS
.
,
.
1
DA
1
M
i
.
'
.
.'
SIN
.
T
1
-
.
.
.
.
.
.
.
r
1 :
1
..
.
1
.
4
>
.
1
.
.
.
i
'
-
.
*
.
.
AR
.
.
..
ET
...
.
-
.
C
.
1
.
C
.
N
.
S
.
.
1.
*
"
.
9.
.'
1
.
2
1
I
'
L.
RY
.
.
.
the
*
.
:
1
3
.
.
21
19
A
.
*
T
.
.
NYU
,
LV
.
within
.
.
Ł
o
12.
Am
AX
.
.
.
.
-
-
CES
.
i
17
TH
*:
F
rii:
.
.
'.
-
i
RINI
t
'
.
.
'
;!
A
12
1.
31
M
-
2
.
!
":
.
V
.
.
"
'
.
.
STO
r.
TO
7
P.
.
a
C
·
7
.
1
.
.
K
UL
*
12
.
.
.
UNITED STATE
.
4.
VO
.
-
.
.
.
-
Fig. 1. Reference Design for a Launch Package Based on the MPRE System.
-
-
ju
-----omgano.com...
1
A
N
LAN

Aluminum Reflector
Gondenser : Tube
.
-
Copper Fins
Fins
.
.
.
Fig. 2. Section Through o Finned Tube Fitted with Aluminum
Refector and Shutters for Installation in a Cylindrical Drum-
shaped Radiator.

.
it
ON
online
::
s
$
+
'
.
Nyt
Fy
:
*
.*
.
.
.
+
.
.
X
.
-
TI
V
A-.
T
*
*
*
7.
*
G
11,
Itin
.
YA
1
!
-
+
-
.
.....
11
--
..
.
.
.
.
.
.
T
.
-
-
1
$
KAN
7
LE
P
.
.'.
.
K
"
si,
'
..
.
.
t
.
.
SY
.
!
i
I
.
.
.
I
.
2
. .
.
*V
.
1
+
.
th
.
I
.
.
.
w
1
.
...
Christian
!
War
1
....
.
r
en
.***
-
i
1
Ali
.
IT
artean
A
.
.
.
517
2
!
President
ALAM
What
r
La
wao wa
..
.
.
.
e
.. twi
.49 Mdo.Then ing at way
*
.
.
.
.
paper
-
WEB D
.
EMUM
-
NUM !!
1
.
*
RAMY:
1971
1.
9
.
!
.
CA
.
.
,
S
i
.
-..
:
.
:
Ruite
Fig.
rect Condenser with 144 Tapered Tubes in 12 Banks for Use in a Potassium System
f Intermediate Size. The unit is shown in the shop while it was tilted 45° to
inspect the installation of the jet pump.
*.*
*
1
"
99 E 21
0
1
9
S
D
1
E
WHISNES
0
2
9
S
0
2 1
ONE
MOI
001
III
FIG. 4

TITUTI
II
TID
DU
DUI
POUCO
NIMI1
001
NICO
0
III
CUIUIT
UUTI
UIUI
M
mmm
XI
10on
mm
O0N
PIAN
INTRAMWINDMIN M00W
NIIMOOONIMIC
TUNUD11011
001002WOON
LUULUULU
DAWNniotinn! M
IMITADUT
1110011111111011110IMUI10011
UNIDODTUNI
NHWtnTTDUUDU
CUTITUTII
UV
OOO
DULU
2010MILINDRI
JUNIO
INNO100011
N10010110
TUORODD
LUUUUUUUUUUUUUUIIIIIIIIII
DichVORIMITI
WHITOOLIDIMIONOU MNI1 DDD
DIUM DIONI
D
ITULIUOTOD
00000000D
12.ITNODIT
NIINI
In0
Dia
MO
0.011.a
MILIH buti
DUDU
0000000
Const UW0
11
DIUI
• sənguxədmət zuTS 20əн Jo səxəs
B_pore SUOTF7 puog PBOI-7IBU DƏ JUL JOJ E TJ JO JO7B5PBT 247 ..
Euy aoTanq1sta ən4BədшƏL IBTxy əца Jo səqвшI2S3 TBƏTartBuy •
D
NH
ulot
UTC
UNIM
AD
VOTONTA
NIIW
WWPOIDUD
DID
MITUD
01W OODO
MIDWID
000 RONINIU
0DVITITTO
LUDUTTUVINI
DMONTO
V
.
DOO
DOLUTION
MOOTTO Li01
ADDITIVTT00
10DITOINT
DIIDIDIITTI
MIINIT TOONIT.
WIRDIIDIIDIITTI
AMIOIIIIIIIII
KOTIDIDUTITI
ITINOTOD
W
IRUITTON100K
RRDDIIND
INTITOIDOINTI-
DUOTIDIIDIINID
01.1100111001PS
M
IMITINDO
III
TODOS
NINI NIITID
WIDOWIIUTT
DUIKI IM000000
NUTRITI
HUO
DONKUTOMOTO
DODIA KININ
DIBONI 100DNIINIKO
DIADEIODIMUL
WIDOWIUDIDO
0
000OTWOOD
10mm
DomTM
100001 011000100
N1001
It
MM
UNT
11
UND
MIMI
LI
OMNIANIMDI
DITO01
WIIT
tottu
Inn
TIL
INTRE TIMIDO
ADITIONS
00DDOROLO1000UIL
INITN0
N
1100D
UDOTIDIT
MONTT111
MITT001
MnNOLOOT
WIN20TOUR
ROTUNIIDUDIDRID
UDO
UU
CO
U10111
INUII
010
U
UUUUU
INDON
Munn
1 II
savivendarQON
Nt0MMITTI
ImunninuIuB
IDIN
III
1000DWN
101010
D02NDITU
Onm000
RM1000
KODNIO
UNIT
00I10101
IMI
DO
DONO
UOLUL
Il
JU
JUDOUINCE
ptuen tim
LLOU
00
12
II
ODINI
N
ORONA
ODNOMOR 10
2011DIIDII
muona
NIJOTUno
111
UU
UJDOO
UUUUUUUUUUUU111111111111111111
1000W
INDOORI
O
ne
D
0DLI000
MILAN DODIDO10UTVI INDJUODOTTI
LIITUNUT D000001111111101111100011011
1111UURINDAMONINIS
ILON
D1G1N1001 100011001
100001IIDID
IDRUO110110000DIDIN0011100
UN
NULDU.
JIONW110001001000RUITI10121 UDD1DIN
1
MUDOUNIDAD
JOONNNN00W00DNIMININO
MIN
menete:20MOMITOIMINIMODO
LONDON
LIVUDILO
DONOIDIDUDUTOILUNI
PUL QUOMODIUININO
UUTINI 1DNDODONTONOMIOIOONILTOUUIL
MODUL
MIDI
JOHN00001003110.000IDDITIONIID1000DOUNWI0110 DOOITTIMOIDON
HM011010101 0001no2IN1000II 100LROUN 1011ent01001111011111111111010
10000101010100ANTIDOIOIOIOIMH0ODOWE WONDOM NYIMIDD000020
DUIDOIDO10TJTOIL0010000111100101100u0MI10000100NIN INDUNUTNOUT
IIIIIIIIII
I IIIIIDILINDUNOVI TIIMIDD0 IPOHILIDI1IIIII
QICHI1000010101111110111111011000111000000101000101010101010
N11000TOI0011111nIDINCOLODJ000000001Int00DINMOP00TLJUN
E MI
ONUNU
N NITTOIOITTITTI1ODUTOTT OMW
NJUOPILOTPRO
UN
DI
D IOUUOITUUTTHUOMIODIDO
TRITOIDONJI0OPINIODONTIJU
00DIDURIIIIIIII10IUDHOLHIDO01100011 DIO
OINT0101H000110111itto.it ,
00
0 DDONIIIIIIIIIIIIIIIIDIDUNT 100001
OIIIIIIIII00011111111001001
JU
100TOILIHIN 01101011001000TV0011
D
ITIOTININIONIT
DU IDIUIDID TOONITUDIN
0000010110 ITTITUDIA
IDIN TI
INDIUIDUTINIIDUDILO
0001
TUDIIDI
UDRIUMI
D
IUIDIUIDIUUIIIITTUUL ODI
U1I110TUTODIUJIUNUIDONUN
000OJIN100111102T100001100D TDIDINDINOP
ODIUMINIUMI
UUTISOINTITOITUT10aW
1000MDULI LONDONILTAL
O
DONTONDITION
DOU000 II
ODDAM
U
100001NnI0ODIn0011010001011011111101
1110001111111011001001001 IIIII
ADIDI
NIUUDIUUDICIUITATUIDIT
WIIDIDIITTIDIINIUUUUU "ADDICO
PIKADU0100001IN000DIVIDIDUNT DITO
nu ININTUIT
UDPUDOULUUVIL
TIMIUI
DUUUUUUUUUUIDUUDILI
HOUT
O0DIDIO
DIDUNUDUDIOMON
TITUTI
Intm001100101100
0 011 OULUD
DIDINIDD
I0200
TIH
01100 MUIDADOWRUITWUUL UMIDI Coman100mmuni
01 DODILOI10WINDOW.
000 tu!
TUTU
OnIDOS DOLO DMITIT
LOW DUINOYNAD1000 UuUL
You COIDO11
DTM0N
C INTUITUITNODIONDITION
VOOL DINNILIIT
100110
10101000101IDO101110IM001010011010101010
JIE
TIMO
100 UU1 (nUIn0001001100110001100DPIWIRDIRBIMOUSINI
IULIDILIMIII
1000TOITOIM0000DIN0001100101000:10UMIINIU
IDIOTILDITOTTI
MIR POL OLIVITI 000000000001nOUTIUTTIDITU DIIDIILID
Pont 0001110MIUI
QUOTITOA
000000001TUDIITOUTUD
U1001NOIRO
In2OTMOTTO
110.0001ITUI101
DIIDII 110DNOTINI0010100111111
I
DDIDDI
IIULITINDIUIDITUTUL
DOL
OLO 001011001010011111111II
ITUTII
IDIUINIMOND
ODILI
Om 1000mm
11101101001
IIDUDODUOD LINDIDIOVITITOR
WOODWIDOCO10000000001 11I0IJI POWIU1110010000
20130100100
MINUNILADWOL
1000000UI NOUDU
NRI900110011001000
DIRRINTIDIU
01100011 IWDIU1100110PPP011UDIOVIDIUI8001000RIDA
CITUDI
110100010000000000
DH'ITUNNUD
UUDIUUTTUUDTUD
IO
DUULUI DIRIT
DITUOTIDID
TOUUUUOO 1T110
TUDIDI10101010110WIDEOIDI
011 DI
101 1
00 PIIRI
DIDDI1IL00DNODUNI
0111t101001111111II1II
1000
D00
0000DM001
On 10 WILDIRDI ULU Dn001IOUIDITULO
LUI
UDUD1
CLUDDICT UUDI
OtoIDO
inuin
ODTUDIO UDUTULDU
ODNOTOWODU WIVIL
UDI 1000DIDI II IIIIIIIIIIIIIIIIIIIIIINIDO10
UUUUUUUUU
TUONDONI
LA 1 DIWDIIDID DI
MINUIATMODITUDIN
O IDM
ODINU DITUTUU
D0
WIvonna OM.
PULLO
ADAU
ILOMINO
DOLINIITITIUDDDDDDDDDD
ODVODOVODI
0
UU
DE) DOLIITUIUTTUU 1000TR1111100110
TOD11010101110110D TRIPLIIDID IITTILIO
000IDOOUUOOUI
DOUNIDEILD
LOLLIUDUDIDUUULDUITUUUU UUUUUUUU
OUTDOOR
UT
JOUDUINO1
QONUNINN101NITORIN1IONOMI rani JUULI
ID UTILI
DO JUUUUUUUUUUUUUUIN
LII
EIWILMITHIUDIONI HUDDU
TUNNUNTAI
UIDDITIILOT DDITUTUMNI
DIDI
13UIRIIUDI000BWIRD
TOUUNI
WIIDIIDID
DHLIUILDIDDIUITINUTNO
1000
210DIN WINDO001100110000001
OLIUNTIMOTH
H100
WINDMAIDD
W1000RNUDITII
ONNI
t1 100TBOLJIDHNINIOWA000000
U MIITUMIVUITOR
C10
OTO0001 VULJINALMN CIT
WIDIDUDULLUM
OLINE
WWII
JUUUUU
UITIUDITUDUDIDATI
n011101100101 INDO0100ml DOIDO
11011100110010100000000001ISLU 1111:111010000 WAIIDI
VN10201L00101NIINIINODIJBROTINIU
11
DIIDII1000RODURIIDILIDIOTARILII II
111011 DILNIUUDIO 1.... U
10 UNIDDIDILIIDUDHIDIWIAITUUT
VIIPURI DOMINIOWIUpon
Rurat
LOI001
HAI HIDR
I ODIUIUUIUITIUITTITUTI
01011110 IDIORUL 011000110101001010110111
DHMIT
PO10001001100101011 DUIDINTTIMIDIT
DIDIRUDIIDIIDIDUOTUDUNNIT
III. HI
In IDONTIDADOITUI LUONTINUNIDOUD
20TLITIDNIJINITIONAL00IVO
IRUDOL DILUMINUTITUT DWIDODHIMN
DIRILIONIMI
UDC Qanun O PODLE
IN2oKMITLOPINIO
UURIMITIDODUITUDINUDDNPM
U100001000100IDEIIIIII
LOUUUUUULI D01UUIUDIULUDTU NUTR
ITAALJDW
INUYUUUIUIDI
TLAIDU ImamTM
WIWIOID00000101H00.000
UUTUUDIITUITA
OUR
DITUTIUI OOOHIDIUIDITUD
DOIROHOID PILDI1ODITIIVIMIT
NUDI OULU DU
DUIUIUTI
UDDANNOITUTUI
MILO
WRITION
01011
Napi II
DU ND
ITOTUOTUITID
OIHIN W10OTOUUDIO
ID000000110001IUHUIL101001000RITION
Non
101NUDI LUUIUMTOTIDI0
ODWUUUU
DUIDELIUOTIDIDIITOIDUUTITUIUTITI IU
ODI IRIDOLOI010011TIMUM
IS
D1RWATT100DMI
N
IUTTIIDIDDONUDIW
LL
020101011DOTTI
DOUWDit
ILIONDOLI AIUNITOU OUVIDOTTO
OUUUUU VOORATOPONNUM
NO
JUU
LUULU BOEIIIIIII
1
1111IDUITDIMO TUDIO
DIIDIIIIIIIIIIIIIII
THIUDDIN 101 00001011111100110 100001010000110010011 00100D
10INDIDIKUTIITUNUIUI
SINULT
IUUDIIDIIIIHDOIMIIIIIIIIIIIIIIIIIIIIIIIIII
ADDTOTO
IN00000000 UNPINDARIUI0ORDINU101110110I
III
ONKO NMD1 D
IUCU0000000IDRUPALMIROIDIDUNTOLDI 1000WIIIIIIITORIN T
OIT
DIIDIUM HII VII
OLIVUDM002001200IMUNIIDUV10Binti 100101PROINIO
TITUITAM
DUITUTII WCIT
UUUUUUL WINNOTUITIROIDIIIIIIIIIIIIIIIIIMIINIIIMIIIIIT
M
IDNUTOUHU THOMONOIM00000001011110101110110110101WINNIPILITIN
MINIMO
MODW001MOBILE n
9000OOONIIIIIIIIIIIDIDUDUILLOU0DIONI 100
NI
2100MKI DUNIDOOLIDO On
BIOTIDIOMAJANDUOLTUURIPIDIUDIO
THIRINIONI
DREnvater DIVIDIO
UlmD1110 JODIDIOT
000tomm'
1DINTIIMTITUUU
2011MD000 UUUUUUUUU
DINIINTHIOCUIIIIIII
IDII11001100011001101DUOD01 OTO010
PUDUCUDUNUDUUUUU
marmo
D
INCIDUNUTUD WINCIDIDUDUD
LULUULUU
VIINI
DUOTUOTTUNO
D
I 100OUODU TUUDUURDUUUUU
IDIOUUUUU
01100100
NUU1IN000110100 VIDONTOIDOTTUUNTOUTUU
ilUUUULIINILI
XIII
Non
DIUIND INDIVIDURIODDODDITION
100DDINOV
DIDINKD100110110WIDILT - VND
I MIPILIIN100 0001pn3I1000000
ILI TIM
III
NSR
, DOI 1010101010001InoNN
Woman
WNURIDAD DOO
TTTINONIMEO
WINRUNDTUDIUNTUUD
000001IINNITLUID
IB011 DIIDIITATTUDIOA
Uw1000000
ULDU CUJUDI 00011ITIPLIN N
DIDADDONI
D
IIDIITTI DO
MOTOCTIT
20 TUTTI
INDIIDI DO
MONITO.com
100 DIN I10
Ianirim
TON
DUULIHUULUI
OOONOODOU.
UDDIDDY
TOODOOD
HOIDIMUM
DUDUJT 10001
duit
1
n
III
POIOVINIO
Rp100
in
NIWER ID11'
00 DCT
1
TIL
ODIO
OURO
10
1
100
100
POJDIIIII 01.11
DIODI
www.macromonoso...
NINO
DIDIN
IIIIII
Rimo
OU
1
D
LION
11
II
IM200.
TOWO01
III
OUDIN
IDOLULOOD
DONDUITTIL
DOBRNMI..
MOTO
AnoIINOMONIDA!
TWITTER
PPT21DIIDIR
WHIRITTO
NDRIIITUONIOIT00
HI11I
III
TUNDU
LIIIDUD
ITINIQUID
UNION...
IINID
MIRIINIT
JLTIDIIDIDUD
DM 200000000
IN
UU
WINDO
Chat
W 01009.
Miti
N
IDAD
ILI
ULLI
DONDOL
DODILDOTUDI
DIRUDITION
ID000WIDU.
DIIVANT
LINI0.101INI
2010
UN
III
.
MIL
NULUIDH
CIRCWO1ROO
MODUL
ODWIDE
N
ITTO
UUUU
ODO
X
-
SeaWEENIS=1VH
UV
DUO
WWE
-
DI
IDIDO
WW
101011
001
WKODULNIHIL
HUNDIDO0IDDIN1IWA
M
ononumNWTDOOWWOOODS
2
III
INOLINO
210
KO
006
H3MOd %01
83 Mod %08
83MOd %OS
FIG. 5
2 900 FTHEHEHEHE
in
o on

700
VOIMITINTI
TOIMIILOIN
TINTINTTT
IIINNIINNIT
S
O400
JUNTO
KE 14.027wemco 490702.
JADE IN U.S.A.
S 10 X 10 TO THE INCH 46 0782
KEUTFEL I OSER CO.
7 X 10 INCHES
TON
DITITIT
V
O
O
Fig. 5. Comparison of Experimental Data with the Analytically Estimated Axial Tempera-
ture Distribution lä the Radiator of Fig. 3 at 17% Load for Three Heat Sink Temperatures
IIIIIIIIIIIIIIIIIIIIII
IIIIIIIIIIIDUMNIOTILISSIMI
IIIIIIIIIIIIIIIIIIIIIII
IUC
TOTAN
TAMIL
IIIIIIIIIIIIIII
IIIIIIMIINII.
F1
*

6. Effects of Shutter Opening Angle on the Heat Dissipation Rate at a constant Radiator Tube Temperature for Both
the Two Hall-Span Shutters of Fig. 2 and a Single Full-Span Shutter.
.
.
.
*.
,LT:;
ORNL-LR-DWG 71036
RECIRCULATING WATER IN
TAPERED CHANNELS
C
.
.
WATER OUT
SOI
Ol!
C331
OOONNNN
SIL
-
23-01
STEAM IN
-
CONDENSATE OUT
COOLING WATER IN
LUCITE
WATER OUT
STEAM IN
Fig. 7. Schematic Diagram of a Flat-Plate Condenser Designed to Permit
Visual Observation of Steam Condensate Flow in Tapered Channels.
F
TV
.
-
-
.
2.
-
--
.-
-
--
.
-
-
-
-
*
M
.
IT



IDT
6
10
4.
:
T
.
-
-
-
-
-
-
71
.
ST
1
-
.
.
..
1
7
.
.
.
.,
.
...
1
.
.
:
SA
:
i
.
s
..
Y
12
.
T
LL
-
mineralne
2
.
W
..:
Fig. 8. Direct Condenser with a single Bank of 12 Tapered Tubes for Use
l.in a Small Potassium System.

ORNL DWG 64-4888
TEMPERATURE (°F)
TUBE DISTANCE FROM
POSITION CONDENSATE HEADER
(in.)
INLET
... 24
CENTER
12
OUTLET
2
3
4
5
6 7
BE NUMBER
8
9
10
11
12
Fig. 9. Temperature Distribution in the Condenser of Fig. 8 when operat- .
ing at a Reduced Load with Potassium Condensing at about 750°F.
On
1
.
.
.
.
VWwW
.
T
.
.
.
21
:
.
T
3Y
ET
NX
US
'
-
.
URU
11
.
ie
1
i TV
V
-
Y
FY
Y
1.
.
iti auto
:::
-
.
.
Cái
Ti.
ww
.
2
.
.
.
(1
)
NI
, *.
.
4
A
• • • P.
2
1
SIA
.
1
*1,
T
.2
I
*
.
'
'
.
.
Ji
.
.
LO. View of the Potassium Condenser of Fig. 8 Showing the Column Buckling Near the Small Ends
of the Tubes (in the upper part of the photo) that occurred as a consequence of the Non-
Inilambransverse Temperature_Diat nihution. Shown in tim. ...
.
2
..:
:
.
:...
.
.
,'.
'T
S
. EA
M...
.
.
.
wewe
-
-
-
-
-
-
-
-
-

-
.
---
. -
*
T
--
i
...
A
V
System of Intermediate Size.
Fig. 11. Direct Condenser with 144 Tapered Tubes Installed in a Potassium
.

i.
.
'
'K
.
.
.
CT
)
1. DIRE
.
.
412
TI
c
-
-
-
-
-
.!!
NAX
YA
A
..
r..
.
PE
.NET
.
.? inson
.1.
17
-T-
-
+
*
*
*
...
-
.
Y
1
LP
.
1
.
'
.
.
zi
.
.
'.
*
.
1
1
AUTO
U
TILILT NAILSEIL MILLID MITT
SAMENLHHHHFHFHOILEREIZ!!
STIILIHEEL WALL
21
.
→
.
.
'
PL
SU
LA
SH
C
1

*
*
2
M
C
.
*
...
US
.
+
.
.
1
.
A
.
1
3
ry
.
1
+
-
Yi
*
TA
3
*
'.
Primera
.
**Monovo
.
.smin
KEMI
.
.
M
IS
.
.
.
1
1
V
.
VY
$
.
1
.
bring
land
12
.
.
i.
.
n
.
spr.
on moments
i
17
.
internet produk
NU
2
.
.
"
1
.
1.
.
*
TI
t
.
.
.
tion den
.
.
.
*
ii.
VID
R.
LE
2.
-..
1
i
SK
.
.
La
L.
T
ST
.
1
.
.
.
-
".
A
.
.
.
It
1,
AT
12
-
Fig. 12. Photo Made with Polaroid Infrared Film Looking up from Below at
the IPS Radiator Running at about 850°F with the Scavenging Pump
Cavitating. There was little liquid in the radiator except in
the form of films on the tube walls.

*
OS
T
CY
+
Y.
.
.
.
.
.
.
1
A-
.
.
Star
.
S
- KZ
,
Y
IN
-
.
!'
in
.
6
.
is
SI
-,
'
'
'T
11
14
4
LS!
-
.
.
.
po
,
1
.
2
.
in
3-
.
.
î
HA
+
1
.
.
I
.
L
11
-
RH
weiter
,
E
".
.
.
.
.
.
-
.
.
VA
*
.
.
+
.
.
A
V
&
::::
.
?
.
Fig. 13. Closeup Infrared Photo of Portions of the Condensate Manifolds for Two Banks of Tubes Made
at the Same Time as Fig. 12. The density of the image indicates that vapor bubbles must be
moving through the condensate manifold and supplying heat by condensation to keep the tem-
perature in that region within a few degrees of the temperature in the tubes.
Atttttttt
+
-
UM
y
LE
T
-
.
é.;
..
.
.
.
.
.
2
.
.
11
.
.
.
2
L
.
.
.
•
.
Tttttt
n
.
At. 1
-
2.
2
m
RE
1
2.
1
.
i
'1
..
..
'
,
t.
=
3
*
*
am
CL
.
i
i
-
L
.
..
2
.
.
2
.
.
.
..
.!
LIT
S
.
-
.
.
.
.
.
.
nr
.
......
V
..
.
.
.
M
uriul
i
-
-
*.
-1
1
+
.
.
.
.
.
?
RY

*
.
S4
oi
.!
Tot
?
'
:'."
:
1
:
.
SY
>
O
.
.
P.
WA
..
7+
H7+7Hz
7+A
is
: :ico
.
1.
ANG
1
.
.
3922
.
.
2
.
A
. Infrared Photo of the Radiator at Essentially the Same Power and Temperature Conditions as
for Fig. 12, Except that the Valve in the Line Leaving, the Scavenging Pump Was Partially ·
... Closed so that the Scavenging Pump Stopped Cavitating, Liquid Inventory Distribution in the
System Shifted and the Outer 8 in. or so of the Radiator Tribes Operated Full of Liquid.
45

-:
IL
1.
HLEHT
11
+++A
1
.
.
i
I
SIA
.
31,
.icoob..
.
.
P
.
*1 YA
IS
.
.
}
.
.
.
+
12
.
i
1
+
.
-
"
.
.
*
2
F
.
.
.
.
,..
F
.
1.
.
TE
toia
STY
1.
.---
.
i.
:I
.2
-*
II
7
ca
.
.
1
.
..
'
I
$
..
.
.
.
1-
I.
.
2
.
, '
'
.'
.
ti
IN
.
.
.m
.
.
-
*
C1
.
1
-
TY
»
.
>
1S
!
YFIR
A
:
.
.
.
..
.
.
-
3
+
.
T
2
.
-
HO
.
/
a
D
EN
V
.
ar
.
E
.
1
.S
A
Y
:
Ti
Ha'.
-
S
Fig. 15. Closeup Infrared Photo of the Outboard Ends of the Tubes and the Condensate Manifolds Made
at the Same Time as Fig. 14 to show the Steep Temperature Gradient in the vicinity of the
Liana Sunless when the Parkban II

!
...
-
-
.
I
.
.
.
T'e, it't.*
2
.
.
.
.
.
"
E'
.
,
.
VISSA
.
*
.
*
.
*
.
*
2
io.
1
--
-.
A
10.
Y
.
-
-
-
F
RE
TS
"
-
-
.
TV
.
ini
WY
11
m
..--
S
-
.
:
.
N
--
-
.
IN
1
M
:
.
$
N
1
.
--
99
.
".
>
a
Fig. :16. A 20 ft diameter Tube Bank for a large Potassium Direct Condenser
Being Fabricated in the ORNL Welding Shop...
'
!
,..
ili utr. 17
ry.
1
ii
PO
2
aina 4
-
{
.
L
.
1
-'-
'.
.
*
.
:
SAR
2
..
-
.
.
e
V
airyr solidarito....am
1
_
M
within
*
A,
LI
*
,
.
'
11
ROTORENCES
1.
A. P. Fraas, "The MPRE - A Boiling Potassium Reactor System, " Third Biennial Aero-
space Power Systems Conference, ATAA, September 1-4, 1964.
2. R. E. MacPherson, J. C. Amos, and H. W. Bevege, "Development Testing of Liquid Metal
and Molten Salt Heat Exchangers," Nucl. Sci. Engu, 8(1), July 1960.
Arthur 2. Fraas, "Reliability as a Criterion in Nuclear Space Powerplant Design,"
Air Transport and Space Meeting, New York, April 27-30, 1964.
4. A. P. Fraas, "Design Precepts for High-Temperature Heat Exchangers," Nucl. Sci. Eng.,
8(1), July 1960.
5. R. B. Korsmeyer, "Condensing Flow in Finned, Tapered Tubes," USAEC Report ORNL-TM-534,
Oak Ridge National Laboratory, May, 1963.
6. H. J. Metz, and M. M. Yarosh, "Experiences and Developments in Instrumentation for
... Liquid Metal Experiments," Fourth High-Temperature Liquid-Metal Heat Transfer Techno-
logy Conference, September 28-29, 1965, Argonne National Laboratory, Argonne, Illinois.
0
.
.
.

montanamorow
.com -**** awam dith more than where the momen
BUK
ALAKKEN
SLK
Meta
L
die
hart
het inte
liten stund es
kleine die hele herlic
h
9/ 8 /
DATE FILMED
END
A:
STYS
SI