HTL TR 1969 067
HTL TR 1969 067
HTL TR 1969 067
-tc q /-3
George Yadigaroglu
Arthur E. Bergles
Contract Nonr-3963(15)
December 1969
George Yadigaroglu
Arthur E. Bergles
prepared for
Power Branch
Office of Naval Research
Department of the Navy
Washington, D.C.
under
ERRATA
Fig. 4.1 Solid lines refer to UNIFORM and dotted lines to COSINE
heat flux distribution
2 5 70
IIIII
III I
Abstract
53 5220
Acknowledgements
""NOPWAN""M InNIM00"pp"
- ------- I - -...
Iwwlmw IN1111
- 5 -
TABLE OF CONTENTS
Pag
List of Figures 10
List of Tables 13
Nomenclature 14
4.9 Conclusions 71
- 7 -
Page
Page
- 9 -
Page
REFERENCES 168
FIGURES
Distribution List
- 10 -
LIST OF FIGURES
Fig. Page
- 11 -
Ig Page
Fig. Page
WINW40
Pill"
- 13 -
LIST OF TABLES
Table Page
C.1 Range and Accuracy of the Numerical Fits of the Auxiliary 187
Functions
- 14 -
NOMENCLATURE
when British units are used were not included in the formulas.
Latin Alphabet
A flow area
G mass flux
g acceleration of gravity
h enthalpy
- 15 -
j imaginary unit
It length
Pr Prandtl number
Plz position-of-the-boiling-boundary-to-pressure-drop-in-the-
single-phase-region transfer function, Eq. (5.50)
p pressure
q"Iheat flux
Re Reynolds number
T temperature
t time
V velocity
v specific volume
vfg v - vf
fg gf
W(s) open loop transfer function, Chapter 8
Greek Alphabet
a void fraction
viscosity
a surface tension
Operators
d differential operator
a partial derivative
A difference
6 small perturbation
sum
average
Subscripts
ac acceleration
b bulk (mixed-mean)
bb boiling boundary
e segment exit
ex channel exit
f liquid phase
f feedback (Chapter 8)
- 18 -
fr friction
g gas phase
gr gravity
in inlet of segment
liquid phase
sat saturation
1 single-phase region
2 two-phase region
Supercripts
dimensionless quantity
peak amplitude
-~-~
lilllMillnbl
ildllel I ~~ ~ ~ g ,Il||11,1 unlII,
- 19 -
Chapter 1
was given to these problems in the last fifteen to twenty years with the
advent of high power density boilers and boiling water reactors (BWR).
developments [1-6]*.
phase region. A model for the dynamics of the two-phase region is devel-
*
Numbers in square brackets designate references listed in pages 168-174
- 20 -
example the well known Helmholtz and Taylor instabilities, bubble collapse,
flow channel. Acoustic waves, shock waves, and critical flow phenomena
the scope of this work which deals only with one important class of single-
to specify the pressure at the inlet and exit plenums, as flow changes
in a single channel will not affect significantly the total flow, and
channels.
[5,7] have finally brought some order and clarity, and, most importantly,
classical example of the first type. Stenning and Veziroglu [7], in their
for some reason acquire a major importance under unusual conditions, for
when the pressure drop across the channel decreases with increasing flow.
of the system, and they occur in the low exit quality region only, where
work are due to the multiple regenerative feedbacks between the flow rate,
the vapor generation rate, and the pressure drop. Inlet flow fluctuations
11lllimbI
IMMURE
IN
MIN
1111161
- 23 -
reach the boiling boundary they are transformed into void fraction
perturbations that travel with the flow along the channel, creating a
total pressure drop is imposed upon the channel externally, this two-
for the generation of phase shifts. This led to the suggestion that
feedback mechanism, Neal and Zivi [5] named them "flow-void feedback
instabilities".
present case, for constant heat input, the total energy contained within
the channel varies according to the mass flow rate. At high flow rates
little energy is stored; when the flow diminishes the amount of energy
storage and release mechanisms, that can create additional phase shifts.
regime changes, the relative velocity between the phases, etc. Although
by the boundary condition that the pressure drop across the boiling
main incentive for the numerous two-phase flow stability studies. Although
it was soon realized that these phenomena could usually be accounted for
States in the late sixties, operate at higher pressures and have overcome
heat flux ratio. There is, however, no assurance that new, economically
improved designs will not bring back the concern about the thermo-hydrody-
between the open lattice core and the single tube channel. It is suspected
glass tubes provided the necessary experience with such systems [27,28].
Early stability experiments [29] gave very valuable indications and guided
background information.
Crowley, Deane and Gouse [32] reported the first experimental results
Although the facility had three vertical, identical parallel test channels,
thus the system could be operated at any point within its range. This
tional space.
be gathered.
transfer heat to the test channel. The use of easily controllable elec-
the test channel dynamics with the primary heat source dynamics.
The effect of the axial heat flux distribution has been recognized
of the channel.
mode to mode inside the unstable region was observed during the experi-
equal to twice the "transit time", i.e. the time required for a fluid
stood by considering the fact that the perturbations travel with the
flow and that the inlet flow perturbation is approximately 180* out
established that these higher modes were associated with the presence of
This is, however, impractical in general, and the models must be special-
The equations of the present problem are space and time dependent.
methods used to eliminate the space and time dependences and arrive to a
stability criterion [37]. The models are termed lumped if the space
some approximate space dependence. They are called distributed when the
01111,
- 29 -
linear fashion.
Although the notion of the boiling boundary, i.e. the point of the
In the first, the models, in taking into account all possible effects
calculations.
phenomena.
loosing touch with the physics of the problem. The stability model
of the channel is then tested by oscillating the inlet flow around this
frequency. For these calculations, the single-phase region and the two-
tional fashion; the equations are linearized and solved in the frequency
enthalpy with pressure, at the boiling boundary are taken into account,
The importance of heat storage in the channel wall and variable heat
transfer have been recognized and the dynamics of the heated wall were
Section 5.2.3, the often neglected wall dynamics might play under certain
- 31 -
boiling boundary and the variations of the single-phase pressure drop when
the inlet flow is oscillated. These two are then used as inputs to the
the wall are less important in the boiling region where the heat transfer
equilibrium between the phases was assumed and, although the treatment of
The various existing models that deal with the dynamics of two-phase
flow are oriented toward high pressure systems as they do not take into
by using a reference pressure profile which takes into account the space
(Chapter 8).
The stability model does not make use of the various semi-empirical
form
6Ap 1 + 6Ap +
2 0
where the two terms are the pressure perturbations in the single-phase
put into a more convenient form for use in the stability analysis.
effects that could not be included in the theory, for example lack of
Chapter 2
EXPERIMENTAL APPARATUS
discussed.
The equipment evolved from previous designs [29], and was designed
with flexibility and simplicity in mind. The vital part of the loop, the
pressure drop across the heated channels. The test fluid enters the
upward. Use of glass in the loop permits visualization of the flow, but
designed for Freon-113, although other test fluids can be used. Freon-113
was well suited for the test program since it has a boiling point of 118*F
produce any exit quality and even superheated vapor with modest power, at
- 35 -
Figures 2.1 and 2.2 are schematic diagrams of the loop. Figure 2.1
Fig. 2.2 gives exact test section dimensions and codes important
components. The basic flow system consists of: the pump, located well
below the entrance of the test sections to avoid cavitation; the main flow
lines; the pump bypass that permits isolation of the test sections; the
total flow metering lines; the main (HXl) and test section (HX2) pre-
the condenser and after-condenser; the downcomer; and the precoolers used
to lower the fluid temperature before reaching the pump. Glass was
The flow is divided between the pump bypass and the test sections
condensible gases from the main condenser; the idea was to maintain the
sufficiently constant exit pressure when the vapor generation rate was
varied during the experiments. For this reason, after an initial degas-
sing, sufficient cooling was provided to the main condenser to keep the
pressure at atmospheric level (see also 4.1). The condenser volume and
The flow area in the test section preheater HX2 was large enough
the range of flows passed through the heated sections. There was,
however, an inertial pressure drop in this leg (see 5.4.1). The constant-
Table 2.1 gives more general information about the loop. Table 2.2
reference.
Table 2.1
- 37 -
Table 2.2
glass tubing. Seven test lengths, each 15-in. long, stacked and coupled
with 0.5-in. thick coupling plates, comprised the heated portion of a test
section. Figure 2.2 gives the important dimensions and Fig. 2.3 shows
the details of the coupling. The three test channels were connected to
each test section to bring the channel exits above the liquid level in
the riser. The liquid from the riser was directly drained to the down-
comer (Fig. 2.1), while the vapor was passed to the condenser. Struc-
tural rigidity and alignment were provided by three metal rods running
Two instrument taps were provided in each coupling for each of the
well.
a second fluid was used to supply heat, the heat supply was essentially
losses) and easily controllable. Each Variac was powering all three
test lengths at one level; the power was equally distributed by means of
- 39 -
1/3 and 2/3 of the total length. The peak to average ratio was 1.15 for
the few tubes measured (1.33 if the unheated length is taken into
wire braid power leads coiled around a heavier coating at each tube end.
The presence of the couplings, the coupling plates and the power leads
in Table 2.3.
Table 2.3
Conductive coating:
thickness. approximately 16 x 10-6 in.
optical transparency 70 percent
total resistance 100 to 200 Q
- 40 -
portions of the loop. The maximum heat input to the test sections is
heat flux. The mass flow rate was limited by the liquid evacuation
Table 2.4
- 41 -
2.2 Instrumentation
in Fig. 2.2. Channel C, which was generally the only channel used, was
ducers, and thermocouples) were the same for both the steady-state and
channels) was used during the stability runs to record all the rapid
transients.
condenser pressure (P95) and the pressure at the bottom of the test
the level of the liquid in the graduated bypass was also used to determine
transducers were used. Their usable ranges extended from 1 psi to 30 psi.
They had been calibrated individually and their calibrations were often
generally below 0.1 psi, typically between 0.03 and 0.1 psi, depending
provided both the static and the dynamic flow measurements. The calibra-
one foot long, were used to feed the pressure signals to three
Great care was exercised in eliminating trapped gas from the pressure
lines.
The transducer signal was read with the d.c. VTVM during the steady-
- 43 -
Fig. 2.4. This unit had provisions for variable gain and convenient
calibration.
obtained.
flow value during the flow oscillations. The average flow was obtained
from the average slope of the integral signal, recorded over a few
However, when flow blockage, reversals or vapor flowback from the heated
venturi was chosen to minimize the pressure losses and was delivering a
very weak signal at low flow (0.04 psi at 1 ft/s). It might have been
flow direction).
- 44 -
the lower 5 percent of its full scale, where there were significant
the output was also sensitive to pressure signal drift, although the drift
was generally quite low (typically 0.001 psi). It was, however, necessary
to zero the transducer very carefully. For these reasons, the venturi
flowmeter was used mainly at high velocities and in the stable region.
At low velocities, below about 0.8 ft/s for oscillating flow during the
The total flow to the test sections and the bypass was measured by
set of seven panel wattmeters. Provisions were also made for the
was checked frequently during the experiments to detect and correct drifts.
Ordinary thermocouple wire in dry wells was used at other locations. All
- 45 -
across the flow area, was used as an inexpensive, average void fraction
gauge [46] to detect flow regimes and dynamic void fraction variations.
Its operation was quite successful and gave valuable qualitative results.
(e.g. a thin insulated wire wrapped around the tube). In any case, the
Chapter 3
independent variables required are the inlet (all liquid) mass flow rate,
the inlet temperature, the exit (or condenser) pressure and the heat
The code was written with the specific experimental geometry and fluid in
brium is assumed, the relative velocity between the phases is taken into
account, the fluid properties are variable with pressure, and the effects
- 47 -
constant diameter, round pipe and for vertical upward flow. Reflecting
the experimental setup, the channel was subdivided into seven heated
lengths and a short unheated exit section, the heat inputs to each test
and the continuous variation of the quality, each test length was
7
qtnq = .Z q. loss )[Btu/hr]
qi --qi
(q. [t/r
where the losses are estimated from a previous iteration, Section 3.7.
The inlet enthalpy and the total enthalpy rise are given by
h. = h (T ) , Ah = q /w [Btu/lbm]
in f o tn
- 48 -
h.
h in
+±Ah-h(p
f ex
)
ex h fg(p )
The conditions now being known at the exit, the pressure drop is
calculated for each segment, marching upstream until the channel inlet is
reached. The two possible cases of subcooled and two-phase flow as well
as the special case of the segment around the bulk boiling boundary are
Ap = Apfr gr ac
that will be used extensively throughout this work. The (bulk) boiling
boundary (BB) is defined as the point in the channel where the bulk or
extends from the inlet to the BB, and will occasionally be referred to as
boiling. The boiling region starts at the point of net vapor generation
- 49 -
2
Ap 4 AL G 0.14
lfr D 2p
Reynolds number for use in the program (Appendix C). The temperature of
the wall is obtained using the familiar McAdams correlation for the heat
transfer coefficient
influence the heat transfer coefficient; however, this was not considered
this procedure, the predicted pressure drop slope agreed well with the
observed one in the single-phase region, even under extreme wall tempera-
ture conditions (Figs. 4.4 and 4.5). Finally note that, lacking an
models exist for the calculation of the frictional pressure drop in the
was used here because of its wide acceptance and the similarity of the
experiments from which it was derived to the present one. The calculation
Afr 2 2tt
2
S-4 AL [G(l - x)]
D 2
pf
f = f(ReZ )
Re G(l -x) D
if
2 2
tt=
ktt @Ztt (Xtt)
which is not strictly correct for the gas phase near zero quality and for
the liquid phase at low flow rates and high qualities. Practically,
however, it does not make any difference as these regions cover only a
- 51 -
x = 2 (xe +x.)i
iteration.
Apgr [f(l - ) + p a] g AL
a = a(X tt)
was used throughout this workwhere again the empirical function was
given by [53]
- 52 -
with
-G 2 - x( 2 x
B(x,p)
(1 - a)p ap
fg
B(T) -G
Pf
necessary to treat separately the segment that contains the zero quality
boiling lengths, and add them to get the total pressure drop. Although
when subcooled boiling is taken into account, due to smoothing of the void
- 53 -
neglects the effects of mass flux, pipe diameter, and flow regime but has
b.) A correlation proposed by Zuber and Findlay [54] for the "churn-
turbulent" flow regime, which takes into account the flow regime and the
a =
X- 1.13[ + 1 -x] + 1.18 g(Pf P 1/4
p pP9 Pf G P2
Pf
c.) The slug flow model proposed by Griffith and Wallis [55], which
takes into account tube diameter, flow regime and mass flux,
Q /A
Q + Qf 1/2
1.2( AA + 0.35(gD)
x
x + (1 - x)Pg /Pf
- 54 -
that the model is probably used outside its intended range of flow regime.
range of interest, i.e. for x = 0.01 to 0.30, the first three models
agree closely, while the homogeneous model overestimates the void fraction.
Section 4.5.
in the channel where the void fraction increases rapidly with position.
Upstream of NVG, the subcooled voids are on the wall and two-phase effects
are small. Levy [56] and Staub [57] have somewhat similar approaches to
takes into account surface tension, buoyancy and shear drag terms, while
Levy considered only surface tension and buoyancy effects. Both models
have "adjustable constants" which have been evaluated mainly from high
pressure water data, and there is no assurance that they should apply
The two models were compared for the present experimental conditions
W
o
- 55 -
plotted in Fig. 3.2 versus the mass flux, for various heat fluxes. As
but the models agree otherwise. Staub's model with f(3) = 0.030 was used
in most of this work, although some earlier calculations were done using
Levy's model.
postulated between the "true" local vapor quality xtr and the corres-
x = 0 for x < xd
where:
c ATd
x = - < 0
d hfg
(in the present case the Lockhart-Martinelli correlation) using this true
quality.
saturation temperature.
The glass channel was not insulated in order that continuous visual
observations of the boiling phenomena could be made. The heat loss from
the outer wall was small, of the order of 5 percent, but not negligible.
Gouse and iwang [27] under almost identical experimental conditions have
measured these losses and tested the accuracy of the correlation used to
predict them. The good agreemenL they obtained justified the use of
1000 Btu/hrft *F in the boiling region. Then the wall temperature drop
drops added to the average bulk temperature yield the average outside wall
temperature, Tout. The natural convection losses are then given by [58]
- 57 -
with
AT = T - T [*F]
a out room
Adding the radiation loss, assuming a view factor of 1.0 and emissivity
of 0.95, the total heat loss is obtained. The FORTRAN listing of the
the relative magnitude of the three terms in the pressure drop equation,
rate, inlet temperature, heat input, and quality under the present experi-
mental conditions.
3.8.1 Effects of the Inlet Temperature and the Mass Flow Rate
Figure 3.3 shows the variations of the total pressure drop, Ap1 ex,
when the inlet temperature is changed, while all the other independent
variables are maintained constant. Figure 3.4 shows the effect of the
mass flow rate. The subcooled boiling and the heat losses were not taken
pbb, the exit quality, x , the length of the single-phase region, zbb'
and the subcooling with respect to the boiling boundary, AT subb = Tsat ('bb
T are also shown in these figures, as they all have some importance in the
required here since the explanation for the behavior is quite simple and
For a given geometry, fluid, and pressure level, the local differential
frictional and gravitational pressure drops are functions of the mass flux
and the quality only. It is thus possible to compare them in the (G,x)
or the rate of quality increase with length and therefore will also be
The three terms were calculated for a matrix of values in the (G,x)
plane and for three different heat fluxes. Contour maps were than produced
by computer for lines of constant ratios. Figure 3.5 shows such a contour
ratio. Superimposed on these are the three unity ratio curves: gravity =
friction = acceleration (for q' = 792 Btu/hrft, 300 W/test length). They
subdivide the entire plane into three major regions. Gravity is dominant
in the upper left region, acceleration is larger than the two other terms
in the lower left corner and friction is the major term in the rest of the
plane.
The 264 and 1320 Btu/hrft (100 and 500 W/test length) lines, also
shown on this plot, permit to evaluate the effect of the heat flux on the
acceleration term: as the heat flux is increased, the "triple point" moves
chapters.
wmwwww 1116
- 59 -
Chapter 4
gross average heat flux, q" 0, 4700, 9400 and 14,100 Btu/hrft 2
Table 4.1
Uniform 1 1 1 1 1 1 1
Distribution Chopped Cosine 0.43 0.98 1.35 1.48 1.35 0.98 0.43
calculated pressure drop and the effect of the heat flux distribution are
All experiments were conducted with a single channel (Channel C). The
other two channels were isolated during the measurements using valves Vll
for approximately one hour to extract dissolved air. During this period
the main condenser water was shut off and most of the condensation occurred
purge of the air from the system. Then the inlet temperature was set by
adjusting the steam and water flows to heat exchangers HXl and HX2. The
flow in the test section, which was necessary to maintain a constant head
and to purge any bubbles. Valve V13 was then completely opened and the
flow was controlled using valves V10 (common inlet to the test sections),
V9 (bypass), and Vl (pump outlet). Since some of the data were taken in
regions which would normally be unstable, the oscillations were damped out
necessary to supply some cooling to the main condenser during the measure-
- 61 -
system would permit, then the flow was increased in small increments.
Flooding of the riser, which had an insufficient liquid drain, defined the
For each data point, the inlet flow, w, the inlet pressure, p0, the
recorded. The instruments were checked frequently for drift, and correc-
shown that the thermocouples were disturbing the flow and the pressure
opposed pressure taps), they were retracted flush to the wall for most of
in Figs. 4.1 - 4.7. To make the local slope changes in the pressure
run temperature was substracted from the absolute pressure. The plotted
the data.
Figures 4.1 to 4.3 show the dependence of the total pressure drop
across the heated section, p1 - p , on the mass flow rate, for two
- 62 -
different inlet temperatures at three power levels. Both the uniform and
the chopped cosine heat flux distribution results are plotted in these
Figures 4.4 to 4.6 are randomly selected pressure profiles along the
channel.
given in Fig. 4.7 includes all threshold and transition points of runs
D4 to D21. For the PRESDR calculations the channel was subdivided into
model were taken into account, and one major iteration was required for
analyzed below.
Table 4.2 gives the average predicted to measured total pressure drop
ratios and associated rms errors for all the points plotted in Fig. 4.7.
- 63 -
satisfactory.
Table 4.2
exhibited by the data of Fig. 4.7. The only noticeable disagreements are
in the low exit quality, high velocity region (x = 0.01 to 0.05) where the
the channel exit at low exit qualities and sufficiently high mass fluxes.
when the flow is leaving the channel before reaching thermal equilibrium.
region (Fig. 3.5), a lower true exit quality, resulting from either
exit, shifting the entire pressure profile downwards. The same trend is
and lower mass fluxes the flow is given sufficient time in the channel to
predictions are low for all but the highest and the lowest mass fluxes.
moreover, does not account for the mass flux effect. The
predictions, the error being largest for the lowest mass fluxes.
, I''11,
1,jIII11,
1110110510
- 65 -
the high mass flux points (where poor nucleation at the exit
high mass flow rates, made in the previous section, seemed at first to
block having a much larger heat capacity and conductivity, and rather
discrete nucleation centers of the heated wall, only a few inches from
and Staub's predictions of the point of net vapor generation (see 3.5)
were always well downstream of the visually observed values and there was
no way to fit the data by any reasonable change of the model constants.
Despite these facts, the pressure drop predictions in the subcooled region
were good, suggesting that the opposite effects of the subcooled voidage
Figures 4.4 and 4.6 show the effect of the two models on the
calculated pressure profile. Again, the differences are small and confined
to a narrow region extending from the point of net vapor generation to the
- 67 -
the three void fraction correlations that do take into account the
using the homogeneous void fraction model. The respective void fractions
formulas of Sections 3.2.2 and 3.2.3. The frictional pressure drop term
for this comparison and the results are summarized in Table 4.3. The
reported pressure drops are as usual measured from the entrance of the
For the few points tested here, the predicted to measured pressure
drop ratio averaged 0.932 for the homogeneous and 0.992 for the L-M
error were 9.2 percent for the homogeneous and only 5.7 percent for the
L-M method.
the end of a run, the fla was reduced again to repeat a few points as
a check, it was observed that very often these points would not lie on the
due to suppression of the nucleation at high mass flow rates was advanced
D20-194 500 41.7 396 0.196 4.344 0.352 O.9U8 4.088 0.774 0.900 0.978 1.006
D20-181 400 43.0 192 0.496 2.899 0.358 0.937 2.539 0.526 0.942 0.922 U.881
D19-191 300 40.0 234 0.203 4.227 0.132 0.412 3.914 0.290 U.416 0.992 U.961
D19-986 300 42.0 119 0.66. 2.37u 0.237 u.645 Z.17 U.291 U.49 1.1/1 1.U22
D4 -299 300 120.3 387 0.293 1.849 0.06d 2.415 1.314 1.1b t.408 U.961 0.964
D4 -917 200 118.0 281 0.260 1.846 0.267 1.311 1.214 u.537 1.331 0.946 U0.85
D17-296 100 85.3 1,54 0.117 4.103 0.028 0.140 3.687 U.068 0.145 l.U24 U.935
D4 -294 1U 116.6 286 0.120 2.759 u.l0U 0.622 2.034 v.241 U.653 1.0u6 u.846
*
Small differences in the friction terms are due to change of properties with pressure
I''
- 69 -
substantial delay before normal nucleation resumes. The main effect should
at both increasing and decreasing flow rates, together with bulk temperature
lying on the two distinct branches of the curve are plotted in Fig. 4.9
which clearly shows that, on the decreasing flow rate curve, the flow was
superheated toward the channel exit (the drop in temperature at the exit
cycle. Note that it was practically impossible to obtain any data on the
nucleation was abrupt. There was, however, also a very slow time recovery
that would take place in several minutes; points on the lower part of the
test sections until the entire channel voided suddenly, in less than a
parts of the channel reducing the pressure at lower points and creating
further voids.
- 70 -
mind when examining other types of flow oscillations as they might trigger
The acceleration pressure drop depends only on the inlet and exit
The frictional pressure drop term becomes dominant near the exit
of the channel, where the flow attains large velocities. In most cases,
contribution. The net effect will of course depend on the relative impor-
tance of gravity and friction (Fig. 3.5), but it is obvious that for
and experimental results of Figs. 4.1 and 4.2 confirm this observation.
(including the flow venturis and the valves) will become necessary for
the dynamic predictions, later in this work. Using all the available
data from the steady-state runs the following correlation was obtained:
8
Ap = 0.00177 p0 0.2 V1.8
Olfr
The frictional pressure drop from the bypass to the inlet of the
4.9 Conclusions
of the pressure drop in low pressure systems is possible when well esta-
small number of segments, and the variations of the fluid properties with
3%, rms error 6%) in the region of interest for dynamic tests (for exit
boiling was shown to have only a small local effect on the pressure drop.
The heat flux distribution was shown to have only a minor effect on the
Chapter 5
extending from the inlet plenum (station 0) to the boiling boundary (BB),
any axial position. The two sets of equations mentioned above will be
varying heat input to the coolant because of heat storage in the wall.
one obtained by Smets [63] for the case of oscillating flow. However, a
put into a much more compact and meaningful form, and the solution is
Once the wall temperature and the enthalpy variations along the
for simplicity. In the present work, some effects that had been neglected
position of the boiling boundary due to the gravity head, to the frictional
boundary and the instantaneous flow, the pressure drop variations in the
linear. An exact non-linear solution for the coolant energy equation can
coolant is assumed (Section 5.2.4.1). This solution was not used, however,
the wall dynamics into the model. The comparison with the exact solution
will confirm, as expected, the validity of the linear solution at the small
- 75 -
Laplace vaxiable, s, and its imag:nary part, jw, the imaginary angular
Section 3.7 it was noted that the losses are less than 5 percent.
numerical work.
g.) The heat generation and the heat transfer to the coolant are
Z
w
Tb(zt)
q"(z)
0 --- q"(z,t)
d
SX
2
2T(z,xt) 1 3T(z,x,t) (5.1)
X2 a t
millimilism mill
- 77 -
with assumptions a.) to c.) the boundary condition at the heated side is
q"(z)
3T(z,x,t) _ (5.2)
3x x-Ok
x=0k
where q"(z) is the applied constant heat flux. At the cooled side, the
0
boundary condition is
h (t)
-T(z,x,t) hc [T(z,dt) - T (zt) (5.3)
DX k b
x= d
It is convenient to write
6 (5.5)
Tb(z,t) = T'b(z) + Tb(z,t)
2 T*(z,x) 0 (5.7)
ax
can be expressed as
6hc - a 6w(t)
(5.9)
ho ~ w0
c
frequencies [69,70].
_ a6T(z,x,t) = 0 (5.11)
(5.12)
ho
where C = (5.13)
Tb'
- 79 -
2 6T(z,x,s)
s 6T(z,xs) (5.15)
2
3x
a6T(z,x,s) (5.16)
=0
ax
_6T(z,x,s) 6
d [6T(z,d,s) - Tb (z,s)] + E(z) 6w(s)
ax
(5.17)
is
C 6Tb(zs) - d E(z) 6w(s)
6T(z,x,s) cosh( x) (5.18)
d sinh(f' d) + C cosh(- d)
a
becomes
where
*To avoid overburdening the notation, the same symbols are used for
the transformed variables, with t replaced by the complex frequency s.
- 80 -
with
2
T = d/
k
In this section the coolant energy equation, together with the wall
any position along the single-phase region. The general solution, for
any heat flux distribution, will be obtained first and then specialized
a.) The temperature and velocity gradients in the fluid are not
taken into account and perfect mixing and "plug flow" are postulated.
very poor in the extreme case of laminar flow, especially for Freon which
standard methods [58] has shown that 100 seconds were required for the
averaged transfer function to take into account the transit time spread,
- 81 -
profile.
d.) The kinetic and potential energy terms in the energy equation
With these assumptions, the energy equation for the coolant becomes
where P is the heated perimeter, and q'(z,t) is the linear heat input
The linearized form of Eqs. (5.21), and (5.23) utilizing Eqs. (5.22), is
1 b(zt)
- + 36 Tb(zt) = -
q'(z) [(1-a) 6 w(t)
V at az w c w
0 0 0
6 (5.24)
6T (z,t) - Tb(zt)
T*(z) - T*(z)
w D
6Tb(Ot) = 0 (5.25)
reference conditions
dT*(z) q'(z)
ba
dz w c
0
The Laplace transforms of Eqs. (5.24) and (5.25), obtained with the
initial condition
6T b(z,0) = 0 (5.26)
are
b z s)
36Tb(z + - 6Tb(z,s) = -A(z) Sw(s) + B [6T (z,s) - 6
Tb (zs)]
(5.27)
and
6Tb(0,s) = 0 (5.28)
q'(z) (1 - a)
hr*F
A(z) (5.29)
2 Iftlbm
w c
P ho
B c
cw 1ftj
- (5.30)
- 83 -
M6Tb(Z's) q'(z)
o
b
+ K(s) 6T b(z,s) = - _ L(s) 6w(s) (5.31)
z gqc
functions
z
K(s)z q'(z') K(s)z'
6h(zs) --
0q0
[Btu/lbm] (5.34)
lbm/hr
6T b(z,s) o
H(z,s)lbm/hr (5.35)
the heat flux to the fluid. From the conduction equation of the wall and
6q"(z,s) 36T(z,x,s)
x= d
q"zS) P c F(s)
Q(zs) 6"w(s) c H(z,s) + k E(z)] F(s) + C
2
Btu/hrft
lbm/hr (5.36)
6h(z,s)/h w
H*(z,s) E h(=)/hfg
6 -- 0 H(z,s)
6w(s)/w hf
o fg
6q"(z,s)/q"(z) w
0 __0
Q*(z,s) 6w(s)/w q"(z) Q(z,s)
o o
h h* P
fg c - H*(zg) + a] F(s)
c q'(z) ) F(s) + C
0
These two transfer functions are plotted in Fig. 5.2* for a set of
Notice that in this figure, and in general in this work, the phase delay
The waviness of the lines of this and following plots is due to the
limited resolution of the digital plotter used to produce these figures
directly from computer output.
-- 1111mimillilil1iilllll WIN
- 85 -
6y
(perturbed quantity)
hase
\w (flow perturbation)
Re
delay = 2Tr - phase
cies, as the storage effects become negligible, the heat flux variations
are in phase with the flow variations, while at low frequencies, where the
6 6w by 900.
wall heat storage plays an important role, q"is leading
(dpc
d = d2 (5.37)
k k/d &
- 86 -
= ___d (5.38)
hoc h
h
c c
and their ratio
10
T hc
C = = - already defined.
h
wall, while the convection time constant affects the transient heat
transfer to the coolant. The values of these constants for this parti-
At large frequencies, i.e. when the period gets much shorter than
F(jW) + TkW (1 + j)
L(j O) +- (1 - a)
w
0
T PAc (5.39)
- 87 -
q'(1 - a) V 0
0 0 -e
H*(z,jw) w h . ~ -+- 0
o fg j
Q*(z,jw) --- a
consider the wall dynamics but did account for flow-dependent heat flux
6 6 the energy equation. This results in
by writing q"/q" = a w/w in
0 0
the high frequency approximation presented above. Figure 5.2 shows that
that this discrepancy might have been partly responsible for the lack of
Approximation
that when the product dpcw tends towards zero, the following approxi-
K(jw) -
V
0
- 88 -
q'
L(jW) -
w
0
F(jW) - 0
In practice this occurs when the period of interest becomes much larger
than 27T (dpcw << k/d) and 27T (dpcw << h'), i.e. when, for equal
k h c
corresponding temperature perturbations, the heat flux "retained" in the
K(jw) - w + B jW
V jo + l/Th
o
qo a/T
L(jw) -2 [(1-a) + jo + l/T
w h
0
and again F(jo) -p 0. But now the heat flux perturbation becomes
-L(s)
(e-K(s)z
H(zs)
K(s)
q'(z) = q' 1n )
0
q (z)coslne
0
I'
-K(s) z
N
H(z,s) = - L(s) e
[K(s)]2 + [ ]
0
K(s)z
e[ K(s) sin (a 1 - ('-) cos ( 1 ) ]
0 0 0
given by
K(s)z K(s)zJ]}
+ (-) [ e
eI
- 91 -
the single-phase region, especially the delay between the inlet flow
peak and the peak of the enthalpy at the boiling boundary are of
Figures 5.6 to 5.8 show the enthalpy variations along the channel,
1, and 4). Both the complete solution and the high frequency approxi-
mations are shown. These three figures emphasize the fact already noted
that the wall dynamics plays an important role, at least under the
along the channel. For all cases there are significant differences both
mately 2 in. long between test lengths. When these cold spots were
differences are noted at higher modes, when the wavelength approaches the
- 92 -
uniform power distribution, were used for both cases. The power input
for the cosine case was adjusted to give the same total heat input in
Figures 5.9 and 5.10 show the results obtained. In both cases the
however, for the cosine case, the entire phase curve is shifted down-
heat flux, as was done in this experiment, the transfer functions for
the exact chopped cosine and the simulated chopped cosine were compared.
The heat flux distribution of Fig. 5.5 with Zi = 0.25 ft and k = 9.04167 ft
represents the case that was stepwise simulated by the "cosine" distribu-
tion used throughout this work. The exact cosine solution is given as
the dotted line in Fig. 5.9 for a zero-order case. The two curves are
When the test was repeated for the fourth order case of Point D18A-196,
a large difference in the results was revealed. The exact cosine distri-
moll.
- 93 -
that the difference was not due to some computational error, the calcu-
lations were repeated with the entire channel length subdivided into 14,
28, and 56 segments instead of the original 7. The heat flux assigned
the exact cosine curve. Figure 5.11 shows how the solutions, although
cosine result. For 28 segments, the two solutions are almost undistin-
guishable.
The two tests described above, namely the cold spot and the
discontinuous power distribution and the wall heat storage might produce
values (Chapter 7). The effect of the uneven heat generation in the
Solution
The fact that, for a uniform and steady heat input, the enthalpy
t'+At
6V
z(At,t') = V(t) dt = V At - A- [cosw (t'+At) - cosw t']
0 W
t'
(5.41)
entrance time, and therefore pick up the same amount of enthalpy on the
in Figs. 5.7 and 5.8 (curves without wall effect). The same effect is
Figs. 5.6 to 5.8 for the case of uniform and steady heat input, indicates
that the enthalpy perturbation travels along the channel at twice the
at z = )/2, where they are most pronounced, this simply means that the
- 95 -
enthalpy-wise most perturbed particle is not the one that enters at the
peak of the flow. It is actually the one that entered a quarter of a cycle
before and has felt the effect of higher velocity (lower transit time,
lower enthalpy gain) all the way up to z = X/2. The particle will reach
will tend to blunt the sharp nodes. The heat storage in the walls
produces such an effect by delaying the heat input (Figs. 5.7 and 5.8).
A similar effect is produced by the cold spots of the heated wall. Notice
that the delay increases twice as fast in the unheated regions, where the
the flow velocity. Therefore, even a short unheated length can produce
regions is due to the real part of K(s) (Eq. 5.32) or, physically speak-
ing, to heat losses to the wall. The effect disappears when, without the
Figs. 5.7, 5.8 and 5.11, can also produce delays between 0 and 180*, which
are otherwise unattainable in the uniform heat flux case which has the
uniform and steady heat input to the fluid, in order to evaluate the
- 96 -
flow (Eq. 5.40), z(At, t') is given by Eq. (5.41). Taking the enthalpy
gained an enthalpy
h(z,t) = q" At
where q"' is the heat input per unit volume of coolant (Btu/hrft )
,,At* z
h*(z,t) = q'" , where At* -
the entrance time t', will provide the variations of 6h at any axial
below:
where
t
t* At* = At ,z* and 6h* = h P
t
q
and variable At*, and Eq. (5.45) provides the corresponding 6h*.
Figures 5.12 and 5.13 show the results obtained in this way for
oil
- 97 -
6V/V equal to 0.1 and 1.0 . Parts a.) of these figures show the
6V/V = 1.
0
waves in the part a.) of these figures. Due to the linear relationship
of Eq. (5.45), points of equal transit time lie on straight lines in the
wave pattern.
reaches its peak value. The slope of such a plot is the perturbation
constant at small 6V/V , and the points deviate only slightly from a
velocity.
delays; the peaks and valleys of the curves are shifted in time when the
6
primarily determined by the enthalpy variations h(zbb, ) at the
hf, profiles (assumed flat in this case) around the BB, at steady state
6h*
6= -bb
bb dho
bb
profile, while the dynamic pressure oscillations will displace it, both
The Channel
61h0
6zbb d bb dh*
(dho f
dz bb - (dz bb
the superscript 0 and
the subscript bb meaning again that the quantities are evaluated at the
of the pressure drop in the single-phase region and will add an inertial
pbb(zbb) p(zbb) + 6 6
= pfr + pin
6w 5ebb 6w =1.8 Ap 6w
6p =bb
fr 3w aw lfr w
0
z*
= - z. G z* =G - n
6p n i in t in Dt A at
neglected.
- 100 -
Figure 5.14C shows how the combined effect of the two dynamic terms
will displace the pressure profile, and consequently the saturation enthalpy
profile, by
6h0 = dhf 0 0
fbb dp pbb fr in
0
-6h+ 6h0
-z bb fbb (5.46)
bb d dh*
dho
f b
dz bb -dzbb
dhf 6w 3G
-6h* + (-) o [1.8 Ap* -- z*0
bb dp pbb lfr w0 in at
bb q'(zb) dh dp
w dp p dz z'
o bb bb
6
where Shbb = h(zbb ,t)
dh
(-)dp = derivative of the liquid saturation enthalpy
Pbb
evaluated at pb (fluid property)
dp1
(- 1
Czzbb total differential pressure drop at zb
b (< 0)
are given by
Mftki IJ",
I111IMMINIM 4 4MIN 111111h
"I''JI11,1141111"'AIN IMI,
- 101 -
6
6h h(z h*(z*) = 6h + ( dh zbb
bb bb bb bb dz bb
except when the non-boiling length becomes very large and the period is
ft
lbm/hr (5.48)
where
Ap*oz
P (jW) = -1.8 lfr
-
in .
J 1
fthr
0
always be in the lower half plane. The dynamic terms will always increase
will depend on the relative magnitudes of the friction and inertia terms
(Fig. 5.15A). The static pressure term has only an important attenuating
effect on 6
zbb*
- 102 -
5.3.3.1 Remarks
around the BB, were assumed in the derivation of the transfer function
presented above. As this is not the case at the limits of the heated
and unheated segments, the solutions obtained are obviously not applicable
variations will be due to the change in length and to the dynamic effects
dp* wo (5w
= ( d) o bbz + [1.8 Ap* - -- z* 6 (5.49)
dz b bb lfr A in t w0
where
6Ap 1 dp*
P = 6 lz -) ((550)
0
lz z (dz z*
bb bb
is the position-of-the-boiling-boundary-to-pressure-drop-in-the-single-
SAp
P () = w = 1.8 jjlfr 5.51)
(i
lw 6w w0 A
z* z*
in E -
11
where z' and A. are the lengths and cross-sections of the various
1 1
obviously due to the inertia of the fluid in the preheater and associated
rate of increase of the flow (dw/dt) was consistent and permitted the
rig. Adding the length of the unheated region between stations 0 and 1,
z* becomes
in
z = 2.3 ft + z* (5.52)
in bb
- 105 -
Chapter 6
equations that will permit calculation of the pressure drop in the two-
channel.
of the model are formulated in Section 6.2, the pressure drop formulas
in Section 6.4.
6.1 Introduction
Several models that treat time dependent two-phase flow have been
proposed in the past, and have been shown to predict with various degrees
boiling length into a number of space nodes and solve the three basic
equations (mass, momentum and energy conservation) for each node, thus
- 106 -
will be taken here. The channel will not be subdivided by space points,
the example of Wallis and Heasley [10], who appear to be the only investi-
drop.
profile are taken into account. The dynamic calculations do not rely on
flow, to about the transit time in the segment. Since velocities increase
rapidly with quality, it becomes evident that the length of a node can
tion has the distinct advantage that exact solutions are obtained for
for given geometry and fluid, the pressure drop in a segment is a function
of the mass flow rate, the length of the segment, the average quality,
and the pressure, this last dependence being a weak one. When the
the quality alone. In the pressure drop formula the mass flow rate and
at any instant of time as the plane at which the enthalpy of the flowing
- 108 -
enthalpy plane is the boiling boundary at high system pressures (i.e. when
absolute frame of reference (e.g. measured from the heated length inlet)
defined as the plane from which a moving observer sees no net volume
flow. The velocity of this plane is then the velocity of the center
j = (1 - a)V + aV
In the Lagrangian description that follows such planes and the quanti-
ties associated with them will be identified by two variables, the time
t and the entrance time t.,thetime at which the plane was coincident with
boundary at time t. (entrance time). At time t > t., this plane will
equals the net volumetric flow entering at the inlet boundary (i) plus
reference
dz.
j (t) = j.(t) - -1 (6.2)
Gi1 dt
t
- 110 -
tions made in writing Eq. (6.1). It was assumed that the heat supply
rate to the fluid is constant and that this heat is immediately used for
evaporation. Thus, the effects of the heat storage in the walls, variable
heat transfer, and thermal non-equilibrium are not taken into account.
Defining*
q' vf
= A g (6.3)
h fgA
fg
Given the inlet center-of-volume velocity j (t) and the position of the
t
Qt -Qt'
= e e j (t') dt' (6.5)
t.
and neglecting the potential and kinetic energy of the flowing mixture,
necessary to consider time intervals short enough so that the flaw does
dh = _ q'
- j d - [ 1-x)v + xv ] dt
w w w w A f g
= q Adt V + x V ) (6.7)
dh = h dx (6.8)
dx 2(x + vf (6.9)
dt vfg
For a homogeneous flow, Wallis and Heasley [10] obtained the same
Av = A At = Am (vf + x v )
- 112 -
given by
q' Am (v + x v ) dt
Am
In the case of slip flow there are no identifiable mass lumps that can
given above.
Q(t-t.) vf Q(t-t
x(t,t.) = x. e + - (e - 1) (6.10)
1
v fg
h(t.,t.) = h ,
vf Q(t-t.)
h(t,t.) = h. + h (x. + ) (e - 1) (6.11)
i 1i fg i
1 vvfg
f
where
= v + x. V
p. f i fg
boundary,
dz dz.
e i + at (6.14)
dt t dt t tt - .=a
e e e i
Ot -Qt
= e e e j ( ) - e j- (tt)
t - t. = At
i e j iti
e I
+ RC(t ,t )
- 114 -
dz dz. At
e _ -1
+ QG(t ,t ) + j .(t ) - j (t ) e (6.15)
t t
e e
the entrance time t. now being arbitrary, but the residence time,
index (t.) can be dropped from the exit quantities and replaced by the
subscript e.
available:
Given:
MAt v QAt
x = x. e + (e - 1) (6.17)
e I Vfg
Owl
- 115 -
vf At
(6.18)
h
e
=h
i
+ h
fg
(x. +
i v fg
) (e - 1) = h
f + x
e
h
fg
-Mat
pe = p. e (6.19)
i
dz.
j (t) = j.(t) - (6.20)
1 dt
t
Ot -&tI
t-At
dz dz.
e e (6.25)
+ QG (t) + j (t) - j (t-At)
dt = -
t
Notice that Eqs. (6.21) to (6.25) specify all the input information
necessary for undertaking the solution for the next enthalpy-time step.
around t = t - At as follows:
where
dj .
dt
1
tion of small variations of the inlet flow during the time interval.
Introducing Eq. (6.26) into Eq. (6.23) and performing the integration,
dz dz. At
e dt + (e - 1) (6.28)
dt t dt t
t t
be adequate,
MAt
1 e -l1
e (t -2 At e
- 117 -
and
dz dz.
e 1
dt t tt dt
At
-eAt -2Ant' dz.
AG .(t) = - j (t)F (1-e )+ G e e (j.(t') - dt P ij
td~p
1j(t
(6.29)
Approximated to the first order, i.e. using the definition of Eq. (6.26),
dz. -AAt
AG . (t) = 1 (e - 1) p. (6.31)
eidt t
Equations (6.29) to (6.31) confirm the fact that the model is accounting
for the important transient mass accumulation and space distributed flow
- 118 -
x = x0 and at the same pressure level, the pressure drop can be written as
simple expression above. It is the fact that Eq. (6.17), giving the exit
flux that makes the evaluation of the total pressure drop extremely simple.
First the mass fluxes at all mesh points at all times and the segment
lengths are established, using the equations of Section 6.2.3. Then the
total pressure drop at any point at any time is obtained using Eq. (6.32),
boundary (BB), the segments and their boundaries are numbered as shown
in Fig. 6.2. The inlet/exit subscripts i/e are replaced by the mesh
BB
1 2 3 i i+1 N N+1
X1=
A0 x2X
A0 100 Xii 1 XN N+l
X
1 2 1
Eq. (6.23)
AC = j* E.
1 1 1
where
Q. At
E. e (6.33)
and the pressure drops would be estimated using some appropriate model.
results when the fluid properties v and h vary rapidly along the
to obtain the exit quantities from the inlet quantities. Small discrep-
errors. It was found that a much better procedure was to calculate first
the mesh points using the constant transit time requirement and calculate
defined by
-. 1 ln( 1)
(6.34)
1 At pi+l
in Appendix F.
- 121 -
length at equally spaced times t2' 3, ... , tk, ... t will now*, be
considered, the time step being again At. The superscript k in the
equations will refer to the plane that entered at time tk+l* Figure 6.3
shows the progression in time and space of these planes. Notice that all
transients.
k k - dz) (6.35)
ii (dt i
.k+1 .k
k -i (6.37)
At
C = E. jk + B. s (6.38)
i 1 C i i
Q.At
e 1 -- G. At E. -At
B. e Q(6.39) 1
1 02
lines of equal enthalpy
tk+2
k+1
- 123 -
k = jk + AC. k. (6.40)
i+1 i i i
k *k (6.41)
i+1 i+1 i+1
G. 1 (Gk + G. +) (6.42)
2 i+1 i
1.81
0 k
L
A k -k+1 -k]
Apfr . G.\ Ap0gri G.i G.
--Gi kk
Ap k Pfri Gi .
= o ) )+ A - At AC i
-k 2
G.
+ Ap* (6.43)
aciG
k k+1
dz dz) + E( s 6.44)
dt i+1 dt i i 1
k k+1 k (6.45)
z. = z. + AC;.(.5
i+1 = i i
The following three rules specify the time in the variables above:
subscript.
By now the reader will be aware of the fact that the enthalpy
The few limiting assumptions made during the derivation of the basic
equations, namely the assumption of slow flow rate variations and the non-
suggest that the model will probably be most useful for moderate flow
deteriorates with distance from the BB, and reverses into an error at
that was applied to the pressure drop calculation near the exit of the
channel results.
Chapter 7
One of the primary goals of the project for which this investi-
gation was initiated was to produce a complete and reliable stability map
that could be used to test various existing or new analytical models such
the risks of physical damage to the test sections, generally the experi-
deep into the unstable region. Therefore, few limit cycle data are
available.
the present system at four power levels with a uniform heat flux distri-
a) extend and complete the uniform heat flux stability map toward
and
- .iIIII illld M
llkil il ls IlliHIM 1111il
- 127 -
c) enter deep into the unstable region and study the characteristics
in Section 7.2. A short survey of the various ways of presenting the data
is given in Section 7.3. Section 7.4 describes the data reduction methods
used. Stability maps are presented and discussed in Section 7.5, and the
the first three trial runs, eight runs were made with cosine and the
stability runs provided tape recordings of the flow signal that could be
analyzed for frequency content ("TR" runs). Finally six special runs
was recorded. This large amount of data required machine processing. The
permitted the mainpulation of large amounts of data, with any desired de-
Appendix E.
- 128 -
Crowley, Deane, and Gouse [32] have shown that the oscillation
lation due to the inertia of the liquid in the preheater HX2 (Section
channel C. The other two channels were isolated using valves Vll and
V12. Valves V10 and V13 were left wide open for all the stability runs.
For a given test fluid and geometry, the conditions in the boiling
Table 7.1.
Table 7.1
Gross average linear heat rate, q' 265, 530, 795, 1060, and 1325 Btu/hrft
0 (100, 200, 300, 400, 500 W/test length)
- 129 -
The inlet average mass flow rate, w , the exit quality, x , etc.,
fixed pressure drop across the test channels, while varying the inlet
the stability boundary over a large range of temperatures (Fig. 7.1 and
Ref. [32]). In the present experiments, the inlet temperature was main-
tained constant while the pressure drop, and consequently the average flow
was generated, except for the lowest subcoolings. Since previous studies,
e.g. [9], have demonstrated that the stability of the system is indepen-
dent of its history or the path followed in approaching some point, the
operational advantages.
In the previous experiments, the power region from 380 W/test length
up to the maximum safe heat flux had been covered. It became immediately
clear that at lower power levels, with forced circulation, the system was
that used for the steady-state experiments (Section 4.1). The steam and
- 130 -
water flow rates to the test section preheater HX2 were then adjusted,
valve V9 was wide opened and the flow through the bypass was reduced to
a minimum using valves Vi, V2 or V3, and V6. Then valve Vl was
completely closed so as to isolate the test sections, and the pump was
fluid in the large bypass was sufficient to maintain the flow in a single
heat input. During this period the level in the bypass, and proportionally
p0, decreased continuously so that the system was led through an infinite
velocity through HX2 was so small that the Freon inlet temperature was
very easily. Besides providing the necessary low inlet pressures and a
way of continuously varying the flow, the blowdown procedure provided also
integral flow signal (Section 2.2.2), and selected pressure signals from
Occasionally the signals from void gauges (Section 2.2.5) and thermo-
continuous trace of the inlet pressure, po. The signal fed to this
a Statham Model UR5 readout unit. Two calibrated bias signals could
be added to the pressure signal to keep the pen within the chart limits
bypass was constant (except at the top), the slope of the pressure trace
was exactly proportional to the mass flow rate, regardless of the temp-
erature in the bypass. The glass bypass was graduated to permit visual
observation of the level. At steady state, the mass flow rates obtained
venturi readings. With large oscillations and at very low flow rates,
only the "barometric" method was reliable. The top of the bypass had an
irregular shape and made flow rate measurements impossible at the begin-
ning of the blowdown. It was then necessary to use the downcomer level
flow. For this purpose, the transparent downcomer was also graduated and
the liquid level was read and noted in short intervals. This method was
less accurate because of the delay of the vapor in the condenser. The
p trace was compared many times during every run to the manometer M3
level and the gauge P93 pressure readings (Fig. 2.2). The deviations
The three recorders and the manual readings were synchronized with
appropriate time marks on the charts. During most of the runs (D4 - D16)
- 132 -
the rapidly varying signals were recorded intermittently with a large time
reduced paper speed were made in order to detect all the transition points
system the small oscillations are masked by random noise, such a criterion
- 133 -
procedure. Jain [23] states that "in some cases the inception point is
not observed." The knee of the amplitude curve versus some independent
variable can be used in the former case, while statistical methods are
plot the rms value of a fluctuating pressure signal, while Mathisen [40]
defines the threshold by extrapolating toward zero the inverse of the flow-
well defined at moderate and high subcoolings as there was little flow
noise and the amplitude of the oscillations grew very rapidly. (Fig. 7.16A).
signal were made, the analog signal was sampled and converted to digital
defined in this work as the point where the decreasing flow trace exhibits
here. In more general studies these parameters include the ratio of the
2
[52] property index (p f/y )0 (P /Pf ). Moreover, the geometry ratio,
L/D, is fixed. For small variations of the liquid viscosity, the inlet
replaced by the inlet subcooling or, better yet, by the subcooling with
q is the total heat input. The analysis of section 5.3.2.1 showed that
Another parameter that emerged from the analysis of Chapter 5 (Eq. 5.49)
T 1.8 lfr
T. 2Tr G* z*
in in
z* /V T
bb o
the inlet pressure readings, the average flow rate, the inlet temperature,
the power level and distribution, the period of the oscillation, and the
peak flow (the last two from the Visicorder traces), were fed to computer
The heat losses were not taken into account as they were small and
BB: Using the measured inlet pressure as the starting point, the pressure
drop and the corresponding temperature rise for each test length are
the position of the BB. The method is reliable as the pressure drop in
The cold spots were not taken into account for the experimental data
by means of stability maps. Figures 7.1 to 7.5 are typical stability maps
with decreasing flow, data collection progressed from left to right. The
tions ceased are marked by a cross. The deviation of the points at very
low subcoolings toward the left is due to the presence of air in the
system. When the Freon was not exhaustively degassed, gas bubbles started
- 138 -
instability.
threshold boundaries should normally curve toward the right [29]. Such
diverge again with a different frequency. Figure 7.6A shows the flow trace
for such a transition point. At other times the transition occurred with-
out decay of the oscillation, by a mixture of the two modes over a few
bilities alone was rejected in light of the complete explanation given later.
VINN,
- 139 -
existence of enthalpy nodes in the boiling channel, i.e. points where the
spatially uniform heat input to the fluid, these nodes are separated by
wave in an organ pipe (Figs. 5.7 and 5.8). The oscillations were assigned
in the first wavelength (z'b/A < 1), the oscillation mode will be called
zbb
order of oscillation = integer part of -b
portion of the channel should not be taken into account for the calcula-
When the (q/w h , ATsubb) plane is divided into stable and unstable
regions, and the unstable regions are further subdivided according to the
transition points become threshold points for some adjacent region exhibi-
as expected, to the right (Figs. 7.1 to 7.5). For each mode, the period
in Fig. 7.11.
- 140 -
Of course, what was said above does not explain why the system
exhibits higher modes and why transitions from mode to mode occur. This
higher modes the "inactive" wavelengths must be added to the length of the
power levels, as the transit time and consequently the fundamental period
become short, the occurrence of higher modes with even shorter periods is
prohibited. On the other hand, with very low heat inputs and inlet
transitions from mode to mode are probably due to the combined effects of
the spatial oscillations of the enthalpy along the channel, the cold spots
wavelength and the position of the BB change and their combinations must
evident, however, that the higher oscillation modes are not nucleation
instabilities.
order, as defined in Section 7.4. This simple parameter does not take
into account the heat flux distribution and the wall heat storage effects.
Both of these modify the spacing of the enthalpy minima as shown in Figs.
5.8 and 5.10. When the required corrections are kept in mind, the order
the threshold lines of Figs. 7.1 to 7.5. All the experimental observa-
from all points of view. Period versus subcooling plots (Fig. 7.11) were
used extensively. In many instances, when there was doubt, the original
of the oscillations that were not noticed during the initial data reduc-
oscillation.
- 142 -
and the zero-order transition points were plotted in Fig. 7.7, together
with the smoothed zero-order boundaries. Five power levels are represented
on this figure, all with a uniform heat flux distribution. At 200 W/test
regarding the mode of the oscillations in the region between these two
modes, as shown in Fig. 7.6C, with the flow peaks grouped two by two.
Gouse were also plotted on this figure and labeled "C-D-G". As the inlet
pressure was not available for the points on these curves it had to be
inlet velocity - inlet subcooling plane. Only zero order points were
increasing heat input is clearly visible. Figure 7.10 shows the thresh-
old points in the (Ap 1 /Ap, q/w h ) plane. The boundaries at all power
between the flow rate, the void fraction, and the resulting pressure drop.
and its effects on the pressure drop will be determined. A uniform and
The analysis of Section 5.2.4 has shown that, if the flow was
a maximum at time
zbb
At = bb
1 2V0
The ratio of this time to one half the period of the oscillation has been
These, according to Zuber [72], propagate with the kinematic wave velocity
At = dz
2 V g(Z)
zbb
- 144 -
The maximum of the void perturbation will reach the exit after a time
At = At + At
ex 1 2
region are short, due to the large mixture velocities. Other authors
[7,24] have used the same assumption without, however, accounting for the
fact that the enthalpy perturbations travel at twice the flow velocity.
P h v
At = At +t = -- Ah + ,, ln(1 x )
ex 1 2 2 q 1 q v vf ex
fgf
- 145 -
the BB. (The period measured after the transitions were plotted.) All
the points taken at the power level of 200 W/TL, for both uniform and
cosine heat flux distributions, are shown. The experimental points lie on
All the zero-order period data are plotted in Figs. 7.12 and 7.13
of these figures and Figs. 7.1 to 7.5 will show that the points lying off
the smoothed threshold boundaries also lie off the period curves. Given
the assumption that there is a unique possible frequency for each point on
moves upwards at a higher heat input (500 W/TL). At lower power levels
three different power levels with the cosine heat flux distribution.
these boundaries with the corresponding uniform heat flux threshold curves,
shown in dotted lines, suggests that the effect of the cosine heat distri-
bution was stabilizing at all but the lowest subcoolings. The comparisons
made in Chapter 4 showed that there is little effect of the heat input
therefore come from the changes in the average position and dynamic
fact, the period versus subcooling plot of Fig. 7.13 shows that the curves
obtained with the cosine distribution always lie below the corresponding
The enthalpy along the channel, for both uniform and cosine heat flux
identical inlet conditions and equal average heat input, the BB will
be displaced toward the center of the channel in the case of the cosine
power distribution. Consider now the enthalpy delay curve, (2fr - arg6h(z,jw),
Fig. 5.9, redrawn in Fig. 7.15B). For any heat flux distribution peaked
toward the center of the channel the 6h delays are shortened. Figure 7.15B
shows how the two effects tend to cancel in the first third of the channel,
while they add up as the BB moves downstream. Since the boiling delays
are small and approximately equal for the two cases, the period of the
- 147 -
trends are evident in Fig. 7.13 where the cosine curves depart progres-
sively from the uniform heat flux curves as subcooling increases, moving
average flow rate were necessary to bring the system from a stable regime
At low subcoolings the transition from stable to unstable flow was much
a hot wire void gauge and a thermocouple. Notice the phase differences
between the flow, the pressure along the channel, and the void fraction.
The pressures in the boiling region are minimum when the inlet flow is
oscillations are more pronounced in the center of the channel. Notice also
portion of the channel up to the last enthalpy node before the BB does
power from this portion of the channel and provide the corresponding enthalpy
hypothesis.
equal to the test length was chosen, in order to be able to cut the power
Ah = q/w = c ATsubb
is true at a unique flow rate only. Four experimental runs were made
(Runs E03 to E06) with the power shut off from the first 0, 1, 2, and 3
(data tabulated in Appendix E) were very much similar for each run.
_ 7 AT n (q/7)__
+n(q/7)
q* - q AT* =
7-n subb subb w T
0
zbb - n AL
+ n
- 149 -
Two complete stability maps, one for uniform and one for cosine heat
accuracy. The effect of the cosine heat flux distribution was stabili-
independent of the history or the path followed into the unstable region.
Chapter 8
STABILITY ANALYSIS
block diagram for the system is presented. Then experimental data are
used to test the model and the discrepancies observed are analyzed.
of the BB, pbb(t), is used here rather than the pressure at the
6Ap 1 6
pbb - 6P00 =
6
Pbb
The second equality in these equations implies that there are no inlet
- 151 -
Figure 8.1 shows the block diagram of a boiling channel. The single-
P lz, and Plw, given by Eqs. (6.48), (6.50) and (6.51) respectively.
in the single-phase and the two-phase regions are summed to yield the total
parameter, e.g. the average mass flow rate. For the inlet flow oscil-
lating with some frequency W, the system will cross the threshold of
stability whenever
diagram of Fig. 8.1 is transformed into the closed loop diagram of Fig. 8.2.
As shown in this figure, the net inlet flow perturbation produces a move-
ment of the BB, zbb, and a two-phase pressure drop perturbation, 6Ap 2 '
- 152 -
transfer function,
6
wf
W(s) = w
0
for a range of frequencies and plotted in the complex plane. Since most
that the channel is unstable whenever the locus of the open loop transfer
"total transit time", Atex , defined in Section 7.6.1. Figure 8.3 shows
indeed that the period of the oscillation (at the threshold of stability)
MMAINIM11M Ni
- 153 -
mode oscillations this correlation is still true, provided that the integer
2- - int [ z bb ]+ 2At 2 ~2
cal work necessary for the stability analysis. The program, listed in
The required inputs are: the inlet temperature, the heat input, the
mass flow rate, an estimate of the total pressure drop, the exit pressure,
not necessarily coincide with the exit of the channel. The procedure
below.
Knowing the pressure at two points bracketing the exit, the pressure
This method fails to account for the true exit pressure as saturation
enthalpy values relative to the reference profile are used at each mesh
order to improve this situation, the given exit pressure was used to
6.7). This assures that the correct value of the pressure is used around
the exit when the mesh points situated at steady-state outside the physical
channel are pulled back into the channel during the oscillations. Conversely,
however, when interior mesh points are pushed outside the channel, no
correction can be applied. Given the importance of the pressure drop near
the exit of the channel in the present system, it became necessary to correct
drop with the time dependent value of the saturation enthalpy in the segment
extending from some mesh point inside the channel to the channel exit.
This method also provides an artifice for approximately taking into account
- 155 -
the channel exit pressure might appropriately account for this effect.
drop perturbation.
method evidently limits the available resolution. It might have been more
sense, and then determine analytically the phase of this function. For
complete calculation of the open loop transfer function, for one frequency
Figures 8.4 and 8.5 show the calculated open loop transfer functions. The
ature [7,12]. The small loops in the figures occur when the wavelength
been reported by Stenning and Veziroglu [7]. The gain of the open loop
experimental observations (see data included in Figs. 8.4 and 8.5). The
curves cross the real axis for values of the period well below the measured
- 157 -
The two components of the dynamic pressure drop perturbation in the single-
6 Aplw, are shown separately in this figure; the frictional
phase region,
component, 6 Aplfr is 1800 out of phase* with 6w, while the inertial
Im dir. 6zbb
6AP1
Re .w Re
6Apm
2z
Plfr
mately 1800 out of phase with the flow perturbation, suggesting that
the delays and the inertia effects were small. This plot shows also two
component 6Ap 2 w, although the enthalpy trajectory model does not make
such a distinction.
Im
orrection to
Zbb
correction to
|6AP2wi Re
6Ap2
Figure 8.7 shows the adjustments that would have been necessary to
correct the results. These are analyzed below and their relation to
flow and flat radial temperature and velocity profiles. The validity of
the enthalpy variations travel with a lesser velocity than expected. The
ary laminar sublayer to diffuse to the central core of the duct, from
In addition, the transit time in the boiling region is also short (of
drop showed that they were of the same order of magnitude. The exit
will reduce the changes in the vapor generation rate and consequently
came to the same conclusion. They had estimated the value of the coeffi-
would have been required for their observations to agree with their
*
In the progress of the experimental work a short movie was made to
study the space - time distribution of the flow perturbations. A
system of mirrors was used to photograph simultaneously, on the same
frame, the inlet and the exit of the channel.
00101 1
- 161 -
region. Both were explained on physical grounds but could not be incor-
of W(s). When the real axis intercepts the locus of the open loop
these figures are in agreement with the experimental fact that at high
Chapter 9
and summarized here and recommendations for future work will be made.
hoped that these studies will provide a theoretical basis for accounting
equilibrium. It has also a high air solubility and a very low thermal
the data. These groups often contained integral parameters that were not
twice the total transit time in the channel. This quantity is rigorously
defined below.
perturbations travel at twice the flow velocity, while the void pertur-
Therefore, the total transit time is defined as the sum of one-half the
physical transit time in the single-phase region plus the vapor transit
changes in the period of the oscillation were also recorded in this region.
mode to mode were shown to be associated with the existence of more than
one node in the standing enthalpy waves of the single-phase region. The
"order" of the oscillation was defined as the integer part of twice the
and is equal to the number of enthalpy nodes minus one. When the regions
the heat flux distribution on the pressure drop was small, the stabilizing
boiling boundary. With the cosine heat flux distribution the period of
the oscillations was shorter than with the uniform heat flux distribution.
phase region.
It was shown that heat storage in the channel walls can acquire under
- 165 -
as stepwise heat flux changes and cold spots in the wall. The effect is
account both the static and the dynamic pressure drop perturbations in
single-phase region and also increases with the frequency of the oscillation.
flow conditions was developed and shown to have several advantages over
drop. The method, however, cannot account for time-dependent heat transfer
9.3 Recommendations
- 167 -
REFERENCES
- 169 -
14. Shotkin, L.M., "Flow of Boiling Water in Heated Pipes," Nucl. Sci.
Eng., 26, 293-304 (1966).
21. Bogaardt, M., C.L. Spigt, F.J.M. Dijkman and A.N.J. Verheugen,
"On the Heat Transfer and Fluid Flow Characteristics in a Boiling
Channel Under Conditions of Natural Convection," Proc. Inst. Mech.
Engrs., 180, Part 3C, 77-87 (1965-1966) "Boiling Heat Transfer in
Steam Generating Units and Heat Exchangers."
24. Masini, G., G. Possa and F.A. Tacconi, "Flow Instability Thresholds
in Parallel Heated Channels," Energia Nucleare, 15, no. 12, 777
(1968).
27. Gouse, S.W., Jr., and C.C. Hwang, "Visual Study of Two-Phase One-
Component Flow in a Vertical Tube with Heat Transfer," M.I.T., EPL
Report DSR-8973-1 (June 1963).
28. Dickson, A.J. and S.W. Gouse, Jr., "Heat Transfer and Fluid Flow
Inside a Horizontal Tube Evaporator," Phase III, Final Report, M.I.T.
EPL Report DSR-9649-3 (Aug. 1966).
29. Gouse, S.W., Jr. and C.D. Andrysiak, "Flow Oscillations in a Closed
Loop with Transparent, Parallel, Vertical, Heated Channels," M.I.T.
EPL Report DSR-8973-2 (June 1963).
N Mvi i 1-111119114,
- 171 -
30. Gouse, S.W., Jr., "An Index to the Two-Phase Gas Liquid Flow
Literature," M.I.T. Report No. 9, The M.I.T. Press (1966).
32. Crowley, J.D., C. Deane and S.W. Gouse, "Two-Phase Flow Oscil-
lations in Vertical, Parallel, Heated Channels," Proceedings of
the Symposium on Two-Phase Flow Dynamics, Euratom - The Techno-
logical University of Eindhoven (Sept. 1967), also
Gouse, S.W., Jr., R.G. Evans, C.W. Deane and J.C. Crowley, "Two-
Phase Gas-Liquid Flow Dynamics: Part I - Flow Oscillations in
Transparent, Parallel, heated Channels, Part II - Acoustic Velo-
city in Two-Phase Flow," M.I.T., EPL Report No. DSR-74629-1 (Nov.
1967).
36. Meyer, J.E. and E.A. Reinhard, "Numerical Techniques for Boiling
Flow Stability Analyses," J. of Heat Transfer, Trans. ASME, Series C,
87, 311-312 (May 1965).
39. Shotkin, L.M., "Effect of Slip Ratio on the Crucial Boiling Length
in Two-Phase Instability," Proceedings of the Symposium on Two-
Phase Flow Dynamics, Euratom - The Technological University of
Eindhoven (Sept. 1967).
46. Hsu, Y.Y., FF. Simon and R.W. Graham, "Application of Hot Wire
Anemometry for Two-Phase Flow Measurements Such as Void Fraction
and Slip Velocity," ASME Multi-Phase Flow Symposium, 26-34
(Nov. 1963).
47. Sieder, E.N. and G.E. Tate, "Heat Transfer and Pressure Drop of
Liquids in Tubes," Ind. Eng. Chem, 28, 1429 (1936).
58. Rohsenow, W.M. and H.Y. Choi, "Heat, Mass and Momentum Transfer,"
Prentice-Hall (1961).
61. Zuber, N., F.W. Staub, G. Bijwaard, P.G. Kroeger, "Steady State and
Transient Void Fraction in Two-Phase Flow Systems," GEAP-5147
(Jan. 1967), EURAEC - 1949.
63. Smets, H.B., "Nuclear Power Plant Transfer Functions," M.Sc. Thesis,
Chem. Eng. Dept., M.I.T., (Sept. 1957).
74. Zuber, N. and F.W. Staub, "The Propagation and the Wave Form of
the Vapor Volumetric Concentration in Boiling, Forced Convection
System Under Oscillatory Conditions," Int. J. Heat Mass Transfer,
9, 871-895 (1966).
Appendix A
This program was also used in subroutine form (PRESDR) for incorporation
temperature (TROOM), the heat input (QW(I), I=1,7) and the various
gross power input, and subroutine SUBBOL is called to estimate the point
of NVG. The exit conditions are determined and the pressure drop in the
for a single segment. Both subroutines can take into account subcooled
the BOBDRY call is omitted and the zero quality point is determined
along the channel are known in a first approximation. The heat losses
are estimated by LOSSES and the location of the NVG point is found by
corrected heat input and NVG point as many times as desired. Finally,
following section.
problems. At high mass fluxes and heat fluxes, there is a very large
and sudden acceleration of the flow at the bulk boiling point, resulting
specified for the zero quality point. The quality then jumped continu-
imposed, and the calculations were ended when a slightly negative "zero
point" quality was obtained. The last consideration assured that the
pressure drop around the zero quality point when subcooled boiling is
taken into account and the BOBDRY call is bypassed. The problem was
on the number of iterations was imposed. See the FORTRAN listings and
A.3 Effect of the Segment Size and Error Criteria on the Code Accuracy
Test runs with variable segment sizes (the 15.5 inch test length was
0.001 or 0.002 give quite satisfactory results. Even segment sizes above
pressure or temperature were taken from references [41] and [42]. These
following pages. To evaluate the accuracy of the fits the entire thermo-
were reconstructed from these functions ard compared to the original ones.
The resulting r.m.s. errors were between 0.0002 and 0.2 percent.
C F R E C N - 1 1 3 P R 0 P E R T I F S
C __ --
-- - - - - - _ _ - -- - _-- _-_ - -
C ARGUMENTS IN DFGREES F IT) OR PSIA (P).
C
FUNCTION TSATFIP)
C SATURATION TEMPERATURE (F) FUNCTICN HFGF)P)
X = 1.0 / (ALOG(P) - 13.47) C LATENT HEAT OF VAPORISATION (BTU/L9M)
TSATF = - .360185E 03 - .4069271E 04*X + .117441E 05*X=X HEGF = .7058670 E 02 - .9218673 F 00*P
RETURN I + .4283483 F-OI*P*P - .1186130 F-32*P*P*P
END 2 + .1250690 E-04*P*P*P*P
RETURN
ENQ
FUNCTION ROLF(P)
C I 10010 DENSITY (LBM/FT3)
X = ALOG(P) FUNCTION PSATF(TF
ROLE = .1033078F 03 - .2657116E 01*X C SATURATION PRESSURE (PSIA)
I - .3389485E -01*X*X - .83079RIF -01*X*X*X T = TF + 459.6
RETURN PSATF = 10.0 **(33.0655 - 4330.98/T -9. 2635*ALOGI0(T) +
END 1 0.0020539 * T)
RE TURN
END
FUNCTION ROGF(Pl
C VAPOR DENSITY )LBM/FT3)
X = ALOG(P) FUNCTICN ROLTF(T)
ROGF = EXP( - .3267308E 01 + .9181870E 00 *X C lIQUID DENSITY (LRM/FT3)
1 + .3519715E-02 *X*X ROLTF = .1035453 E 03 - .7105255 E-OI*T
RETURN 1- .6448694 E-04*T*T
END RETURN
END
Appendix C
The Fanning friction factor, f for smooth pipes was obtained from
appendix, and the r.m.s. accuracy of the fit is given in Table C.l.
(P2
C.2 The Lockhart-Martinelli Two-Phase Friction Multiplier, ktt
2
The values of 4t as a function of X are tabulated in Ref. [49].
ktt tt
A least squares fit by a sixth order polynomial gave a satisfactory
By definition:
X= gtt
tt ttt kt
2
At x approaching zero, X tends towards infinity and 4ptt must
lim = 1
X *0 tt
tt
x + 0
2
When x approaches the unity, X tends towards zero and 0g now
tt gtt
must have the limiting value of 1. As tt = gtt /Xtt it follows that
lim 2 - 1
X +> tt X2
tt tt
x + 1
subprogram.
was arbitrarily set equal to 0.97 for Xtt values smaller than 0.07
end of the range, for Xtt larger than 100 (x smaller than 0.007) an
Table C.1
Lockhart and Marti- F12LF(XTT) 0.01<X <100 4 values per 0.80 1.53 @X - 0.02
nelli's friction decade
multiplier .2 (x
Itt tt
Lockhart and Marti- ALFAF(XIT) 0.07<X tt<100 4 values per 2.01 4.86 @ X - 70
nelli's void decade
fraction, 0(Xtt)
FUNCTION FF(RE)
FANNING FRICTION FACTOR FOR SMOOTH PIPES
X * ALUG(RE)
FF EXP( - .1805868 E 00 - .8502377 E 00*X
I + .4752425 E-01*X*X - .1134451 F-02*X*X*X )
RETURN
END
FUNCTION XTTF(X)
C LOCKHART-MARTINELLI XTT PARAMETER
REAL MUL, MUG
COMMON /FLUID/ ROL, ROG, MUL, MUG
IF(X .GT. 1.0E-301 XTTF =
1 (((1.0 -X)/X)**0.9)*((ROG/ROL)**0.5)*(MUL/MUG)**O.1
IF(X .LE. 1.OF-30) XTTF = 1.0E 30
RETURN
END
FUNCTION ALFAF(XTTI
LOCKHART-MARTINELLI VOID fRACTION CPRRELATION
IF(XIT.LT. 0.07) ALFAF = 0.97
IF(XTT .GF. 100.0) ALFAF =
10.2661 / XTT
X = ALOG(XTT)
IF((XTT .GE. 0.07) .AND. IXTT .LT. 100.0)) ALFAF
I EXP( - .25435?1 E 00 - .1478533 E 00*X
2 - .3270721 E-01*X*X + .1222962 E-0?*X*X*X
3 + .4128012 E-03*X*X*X*X - .4598919 E-0*X*X*X*X*X
RETURN
END
FUNCTION
F2LF(XTT)
C LOCKHART-MARTINELLI TWO-PHASF PRESSURE
DROP MULTIPLIFR
IFIXTT .LT. 0.01) FI2LF 1.0 /(XTT*XTT)
IF(XTT .GE. 210.0) FI2LF = 1.0
X = ALOG(XTT)
IF((XTT .GE. 0.01)
.AND. (XTT .LT. ?10.0)) FI2LF =
( EXP( .1445175 E 01 - .5039395 E 00*X
2 + .578383 E-01*X*X - .1157590 E-02*X*X*X
3 - .4390361 E-O0*X*X*X*X + .3007357 F-04*X**5 ))**2
RETURN
FND
c C OC CC C
a NC W a 'D0 Otz C4 .0Nft -t. 0 4
.0 a C, aONI
C.4 C
oo a eoocc
IIIL .,
CN N~~~ C.$I4. tr 4-
afte t C NO-fl.t 0-
VIO OCCa CC C C &C
~~~~0
3 -3' 1217
0 .e 0.-' 0a Inc
3tf
0
Of0 - NtU) N I. ON-,. aI
.0
W,0 m mNP-N 01C imn C
Or .
-- N~
yf 4 W -CI4
CL
C0 I-j 3 001 10z400-10
0zzoozzzoo2zozz
0 DQ D -a I
-. D n w D
6 iNu
U.S L) ;- -c m.4 Ne LW f- 4- -
P- I P- CMC C.coc -CellCneCN
COCC 0CC C C CC
.f-.l.
'0 CN UI; N 4N
;.N .
N'C 0L .0 4
4 40D04044' 00000
NQ
0 m0
mC.'.- c I.P-0 c mCC
CCC t. CC C C. -C
Z
I- C; C;CC- C; - -. 0.00 4
!-0
C CC7 C7 4- .f C C C .. ;.. CC
IL C.tNC.N.0CC NI' I40-fN4 0N N
2 f
Olt, 0 C .. t..%N cf
O,*Cv 00 a0NC4K....4N4
0 C44SCNM)N-
r0 - 1
CCC CC C . 0 N 01010.044.40.0.0.0
V.'44V 4 0O P0 WA -C
aOo .. 0'0003 '4
0.0.'A,0348...00 .
- ~ ~ A
. w. ww- 0..3 8,.
3 0 0 - a0 0
0 .V a 0 40 V *0,V
A 0M 4
000e 0000e
42. .. 0.3.N.3W O
3* - -4 Z o w VA
Q3 3NNQ4 O ON0
W-
4 -A 003.*
A --
........ e .. e. CP
4 . .aw. . .- 4 N-0.0-4.4--4
.i ~
- 0...-4N -4'444- . - - a
-3-A a. A
.0.0.0....e,0.e. .e.
i'4Nd0'8N0' 0
z 00. 30'33-4-4-2
0.N430.
00.3.0 0W 4-00 .ooIR 8
a g
0
al i
0 0 0'3 3 '000. 0.0
O.-NO.4NW N WM0O0 3
.d.0d0.ON83e..9900O .3 N3i3
E
N40000.i48.3-NQ44O444440
N4 N3.483N.8-NOOO...QMN3OO.0U 0
N N4
w a -.w 0.0 -a wN a. N # N . NN 9O.0 0..O00 . w a 0.0.u.00 41 0'-4 -84 itN
Cee
, W -A 3 e -E'
NONN0N.0.000".40-aU10.0IN-N
NN e4:s Vs m Am $4 0 CeO. a
2
w I. In 490-.-m
ad
O0 w v-N a 8
-ONNW && 0N& .8e,30000NNe'4 w Oi 3 A . 0 a-4 a . % as .0
W &N. . . . .i3NN
.a*-
' . 0
eW A0 . N 0 . 7*
N--ONN.0N8WWWON NNN43 NON 8
' A 'A 'A A t W .V . 21. v O N
A4A3A z N. v r-, 1.
43%24.02400A-.44404---00.8
JO.00O00.0N0-3 C
V.i O.
A
<0 ----- NNNNMNNNON--- 0 -
4 OO0000000000OOO000O00MOO
o i | Ii i iii I iI I
O4 0N -- 0O .4 00 4N4-.0e'
0Mme
O- 440 e e -404 044440440
Oc a, o a, a, a, el a, a, eY C, a, Oe e ol (7 C) a, 0 a, 0 0- e
4 000
OO 0 0
0OOOO - .
OOO-OO .- 0
.OO
OOO
do i
4 e 4 4. 004
Nm 4 044le
e
x 0 000 444m444444 O00 N
000 Q44 444 NN4 4@404 44Imo 0-
044N444N4444 4NO 0. 4 4 N4 N04044 N
0 0--- 4444 4 ON4N444
4 444404040004ee
OOOO4444404N4044440444404
e 440e ee ee ee ae ee4 4 o44..4N.-14404440400404444 4 4 0
04 00c0c0N O N
0 0 4 4
O N4440 44044 40044 0 N440
0-00 4 4 444~.~ 0 4O4.44 04 44O 444o 4444oM
O 4
Z 440,-04044040NO40 NO444O44
O 00.A -l0 444 40 400eme4 44M0N
0
08o 0
404444440400m40040444004444N44
N m - 0- 0- - 00 0 0 ~------- 0O0
ol a, e l C e e a, el o e e a ,a a, a l m m l c m a a, m l c, o' e a, el 0
0n 04 ~ U0 0 0 t.
NNo
0 0
000 0
C -
61.1000
N -t GOM tt0
N D N O 0 M NOC O
N000 4400.4044M4OO4NOO44NO4404044
0004m4e0taWmeN4044444
.1e o S40444404404
0440444404e 44e4ee4ee00e4044e 44e 4044044oe *
.. 1-N1044 44 00 44440N440NN44m4 N04
Appendix E
S T A d I LI T Y 'L X P L R I ME NT S
d0 SERIES - TRANSITIONS
9 SERIES - FIRST OCCURENCE OF USCILLATIONS
100 SERIES - FREUUENCY OF NEXT UNSTAbLE POINT WAS USED
200 SERIES - THRESHOLD ObTAINED 6Y INTERPOLATION
900 SERIES - TRANsITIONi TO FUNDAMENTAL MODE
U N I F 0 R M H E A T F L U X U I b T RI 8 u T Iu N
RUN-PT OW TIN N PO P1 DPEX XEX OTSUIN DTSUB8 J/WHFG TAU WP/WA OPRAT ZRAT DHAAT DELDM ORDER CO4MENTS
IWI (F) (LBN/HRI (PSI) (PSI)(PSI) (F) IF) (S)
TR2 91 100 121.5 279.0 16.54 17.51 2.96 0.152 5.6 3.7 0.136 4.20 1.10 0.403 0.094 0.097 0.67 0.256 808. HX
TR2A 91 100 123.0 419.0 19.61 18.53 3.98 0.112 7.3 4.2 0.090 4.65 0.409 0.161 0.165, 0.686 0.262 8UL HX
DIZA 92 100 123.5 700.0 21.26 20.03 5.12 0.073 11.2 5.0 0.054 4.00 1.16 0.518 0.325 0.335 0.732 0.368 RUBL. HX
012A 93 100 123.2 572.0 20.43 19.28 4.37 0.084 9.4 4.6 0.066 3.94 1.36 0.485 0.245 0.252 0.714 0.344 8UBL. HX
TR28 92 100 119.2 168.5 18.34 17.34 2.64 0.231 7.4 5.6 0.225 3.60 1.27 0.420 0.067 0.090 0.734 0.456 OSC.DIES
T728 93 100 118.9 149.0 18.18 17.18 2.48 0.259 7.1 5.6 0.254 3.65 1.34 0.422 0.077 0.079 0.7?7 0.448 OSC.DIES
TR28 94 100 118.3 131.0 18.05 17.06 2.36 0.292 7.4 5.9 0.289 3.80 1.33 0.425 0.071 0.073 0.725 0.455
D4 294 100 116.6 285.8 19.23 18.20 3.46 0.129 12.6 8.3 0.132 2.95 0.529 0.719 0.225 0.936 0.827 INTERPOL
DS 932 100 103.6 85.0 17.82 16.82 1.98 0.394 21.3 18.3 0.446 8.65 6.05 0.627 0.142 0.146 0.947 0.622 W EST
D5 299 100 105.8 171.2 19.52 18.50 3.66 0.177 24.7 18.7 0.221 3.20 0.601 0.293 0.301 0.914 1.719 INTERPOL
09 298 100 101.9 165.5 19.12 18.11 3.66 0.176 26.5 20.8 0.228 3.57 0.630 0.314 0.324 0.919 1.719 INTERPOL
D10A 90 100 91.5 47.9 17.62 16.61 1.95 0.699 32.7 29.9 0.790 7.25 0.612 0.130 0.134 0.621 1.220 TAU? PT?
017 296 100 85.3 153.5 19.80 18.77 4.17 0.135 45.6 35.7 0.246 2.64 0.788 0.496 0.510 0.784 4.009 INTERPOL
016 293 100 72.8 103.4 20.20 19.17 4.43 0.209 59.9 50.3 0.366 3.08 0.720 0.467 0.481 0.745 4.854 INTERPOL
018A 196 100 62.5 102.0 20.02 18.98 4.61 0.184 69.1 58.4 0.371 3.90 0.772 0.532 0.547 0.935 4.464
018A 187 100 62.3 83.0 19.37 18.34 3.97 0.268 67.4 58.5 0.455 5.85 0.747 0.433 0.446 0.018 2.981 W ?
020 98 100 44.3 113.0 21.80 20.74 5.84 0.081 92.4 77.4 0.335 3.80 1.30 0.858 0.773 0.795 0.755 6.095 W ?
D20 189 100 44.2 96.6 21.46 20.41 5.51 0.138 91.6- 78.6 0.392 3.58- 1.40 0.802 0.671 0.690 0.794 0.000 END DSC.
D20 90 100 44.8 88.8 21.12 20.07 5.17 0.174 90.1 78.1 0.427 4.20 1.20 0.787 0.613 0.631 0.719 5.565
D20 82 100 45.1 75.9 20.54 19.49 4.59 0.248 88.1 77.8 0.499 3.30 1.40 0.767 0.522 0.537 0.150 7.055 TAU*5.80
D20 83 100 45.2 63.5 20.10 19.05 4.15 0.345 86.7 77.9 0.597 5.80 1.20 0.730 0.438 0.450 0.603 4.022 TAU-2.95
DIZ 98 200 125.1 516.0 21.10 19.98 5.17 0.172 9.5 6.6 0.147 0.333 0.15? 0.162 BHX,THR?
TR3 92 200 120.0 262.0 19.15 18.13 3.43 0.298 9.1 7.3 0.289 3.00 1.15 0.351 0.089 0.091 0.697 0.359
D4 917 200 118.0 281.0, 19.39 18.36 3.62 '0.271 11.7 9.4 0.269 3.52 1.21 0.382 0.122 0.125 0.719 0.392
D6 199 200 115.6 265.0 19.38 18.35 3.51 0.277 14.1 11.4 0.286 3.96 0.415 0.140 0.144 0.752 0.424
06 196 200 114.0 211.0 19.06 18.05 3.21 0.344 14.8 12.5 0.359 3.81 0.414 0.121 0.124 0.754 0.480
05 294 200 107.0 223.5 19.95 18.90 4.06 0.799 24.7 20.4 0.339 5.00 0.456 0.210 0.216 0.831 0.603 INTERPOL
09 94 200 103.2 202.5 18.86 17.84 3.39 0.326 24.9 21.1 0.373 5.40 1.68 0.503 0.195 0.201 0.823 0.576 TAU ?
010A 93 200 94.0 194.5 19.09 18.06 3.40 0.307 34.9 29.6 0.389 2.74 ~ 0.597 0.262 0.269 0.765 1.598 TAU-6.60
010A 284 200 93.8 142.8 18.57 17.55 2.89 0.446 33.4 29.6 0.530 6.60 0.563 0.192 0.198 0.865 0.662 INTERPOL
TR5 92 200 93.3 262.0 19.99 18.94 4.24 0.203 38.2 31.0 0.289 6.4C 1.06 0.631 0.369 0.380 0.869 0.717
D17 197 200 85.4 214.0 19.56 18.52 3.92 0.242 44.8 37.6 0.354 7.50 0.665 0.364 0.315 0.879 0.743 TAU ?
017 182 200 84.9 87.0 17.85 16.83 2.23 0.756 40.0 37.0 0.870 7.25 0.590 0.145 0.150 0.910 0.755
0184 191 200 62.6 440.0 21.35 20.21 5.84 -0.014 72.6 49.8 0.172 6.15 1.051 1.000 1.000 0.707 1.233 W ESTEST
0ISA 182 200 63.0 300.0 21.07 19.98 5.61 0.067 71.6 56.4 0.252 6.15 0.872 0.755 0.776 0.744 0.000 W7ENDOSC
D18A 93 200 63.4 197.0 19.82 18.76 4.39 0.200 67.6 57.4 0.384 2.84 0.774 0.504 0.519 0.427 3.006
318A 185 200 64.0 143.0 19.18 18.14 3.77 0.347 65.1 57.6 0.529 4.68 0.694 0.368 0.379 0.888 1.833
018A 287 200 64.1 122.5 17.36 16.32 3.43 0.453 64.0 57.4 0,617 9.05 0.672 0.314 0.323 0.929 0.945 INTERPOL
DISA 280 200 64.2 88.0 16.54 15.50 2.61 0.694 61.2 56.5 0.659 8.49 0.664 0.222 0.228 0.801 0.990 INT@RPOL
016 97 200 72.4 140.5 19.70 18.66 3.92 0.381 58.3 51.8 0.539 0.00 0.620 0.327 0.337 0.000
016 289 200 72.4 126.4 19.20 18.17 3.43 0.440 56.8 51.0 0.599 8.70 0.635 0.290 0.298 0.934 0.872 INtERPOL
018 290 200 61.4 198.8 20.18 19.1? 4.53 0.187 71.0 60.0 0.380 2.32 0.786 0.532 0.548 0.836 3.856 INTERPOL
018 186 200 62.3 103.5 18.61 17.57 2.98 0.541 65.0 59.4 0.731 4.70 0.684 0.274 0.262 0.913 1.881
D18 184 200 61.8 119.0 19.36 18.32 3.73 0.444 67.9 61.3 0.636 -.-. 0.644 0.326 0.335
020 90 200 42.5 204.0 21.33 20.25 5.35 0.111 92.8 78.8 0.371 4.00 1.20 0.864 0.709 0.730 0.894 2.948 THA ?
020 184 200 44.0 116.0 19.61 18.55 3.65 0.398 86.3 78.2 0.653 3.80 1.90 0.760 0.401 0.412 0.930 3.078 TAU CAHN
020 186 200 45.1 81.3 18.78 17.73 2.83 0.680 82.7 77.0 0.932 6.00 2.50 0.718 0.277 0.285 0.936 1.919 TAU CHAN
z kZ zZZ - JJ 3 22 .707 42
E r 0 W
2
0
T1 ,
< 7
4 w teta
W.
2 20M * e7 N * e e
W N
4 2 tN * U 000
Z7 < << z l.
<N<
0 I+ ae ta
P. N
Nmx 4N
C, ZW 3 0 770707 27772 2-.030
7 47 < 773
a. ~
~~
,c ~~ ~ re I z r cr 7 7 acct @, N- In . 0, 1 wMM 0 Ow2 00r2M7C
N
,,,a Wa 1 O COOc CC) 0 .- 9 N0 0 0 0 0 0 0 N 00 oo 0 -0 -000
ce & e i l @ i i
o I
0 0 0000 000 022
4 Z; M
-z r 01 '-
.. 0r02 0 NC., er 0 .C 0000000000007N 0 4 m N
A 0000000 0 W N 0 nW
0 in0 0C,0 227027322222727
2
0 0002722702727222777
z
00 c, m DCw a(Y
000000
0 a a
or o WU W e e
4 7227200770020770'
mm 4 )r4 4 &M0O4 NM
N (7 a An4
7 22O0732022422770277
,-,.-
- - o4
O222 . w no j No - o. N
. 2. . . ,reDcir<7T
. 2 w i7'e7- 0 C' t' o o M t re
2 322.7@77222222722N - 0% mt oo w
e M w w* 0 M r Io DN
:0 aP 002 o000000000P-e e'1eZ c! It
00oP-fu
o
0,e e
0 Z 072O0c!20- 222P72770 M
O aO O OO OO
077
_j
.
.c!
m r XOOOoCo
4P 4 4 . e . . . . e e e e e e ee e . . . e e e 7-
2 00000000000000000
77220720732222-O
Q
s.. 4 n
cOOOOOOO)OW
O c oOC
,
Oc-O
77O 227OOOO.7 <3 0207700002e e ee - 2 ee
S - -o 'D .- + MW m mn -7 W -o a N W mr O0OP
N <
0770777772
7077c
N
777027' eeo 4o w 200-NN 00300700270202M70
N m P"m " e NP- 4 e4
- 3 .- 000- 002020000.-00000. 0000.-
N 000 0 2 r o re 4 io 4 000000 0-. 70 #- 2 20 77 300022722 4
4
3~
N NTO4
I-N0-0
2 02270 2O 7MO 0. P-2272777202323-7272774 -
0000000
0- 077333733227222-r
O ONn N 0NO N
a, A 4 4
30720N02N777707707
c,
0
; 'N c n o I-
0000000000000000002ee
oo
7727 7:
e e4 n 4&
P'P--
N 00 00 000 00000 0C t0L72O7OOOOOOOO7OO0OOO2
l O r O C) 4 Ot m. N4 WA feN 'b-A
D-0 . 22 2722 27 77 22 2
an a a-
000000 .-
eo a- 0000 00000000 A e-_N--- NOO N4N- NN @ 0
02 207 72
4 -
2o o 2 e e 2a 2o 7a 0o Z 2o 7 4l C,P- eeeeoeee 0. 77772277\7n 702777 4e oP.
-.. CL
'N000 e
2 0e 07N
0-27777 e7 .77
m - nr Oe0 N4 20.72 27 -22 777 204
O777
N 0
P'-0-00N00O
o N * 07727227227207N @ O 2N
M - m l O7
N c
N
l'
37 22
r- I- I~
22
Pn4
77
_ i 4C 4
7
4 inw
N New : Noe oA 700man 0. 77700077772- 77774N ta
m m NN
0 -.
<
2270N730707272O20272
27o..
0 0o0 220,.-22 27,4004
eO M 04 *....7730 e2 ... 7
OE0
0 ... e.2e0
e 0, e o
3 -2 N372 n
7
- - 77-. e .- '0 . -. 7027e 7e e 7 e . ea e e0ee 070000P- ra 0.
0.
N0;; O
OOOOOOOOO000O7007
e o e e
Nm a, ") OCOOOOOOOO220OO
<
In- N M N- 0 01 N @4 P. ' m w
2702704
700270e
4e 00 00 00 0 00 00 0
e...e2e..e. -0
3..
00002
mm202
770270272072 ee ee e22
2002
- S
722N207722N7N77772
70e7e-
eeone comm o
7e2e 32e.... .. noe coeae
2e 7 eee
7 00. . e. ..... e...........
-- .0 w7727 0 --
000000
oe ee
o a, < m Q ~ae. o on o o o o P-o 44 n
277777 -0.L. 0,7722222N 222777NO 7P to
3. .. ... e. .. ..... e22 7 .. 7 0 .e.. e 2... I;, r; 8,; 1; 41 0. 77777e 77722. 722002772 e *
000000e e
300
- 072722727N 3200- 207,..o n72022 P-702N7.72 3oOOOOOOOOOOOOOO
707e e.00 2 7 2.e eeeee 73772072000770 e . e 7777272770
03O00
00000 00000 7 000 00 220000-0
3220.-3020N22772723N7.- 227N
N
0
7 4
N002200727770022
0000000000 Of . e .. e .. . . .. . .e.27
20.
*~
2027702772
0 772
2007 7222
220020277
777 77777N
0000.,
7
2 2727077727727707202OO O O
0000000000000000000000000e00e00ee ee ee .e e 2 720777-7in7772222
03 00000 00000 00000 00000O000 0 00022002200e
OO0O0 00 00
20e0e 0 e
0
e
0.N 7 72.- 7007 0707072702002NN - 2N 00N7- 7- 3
inin 03 0000000000N 000000000 :-- 4
2
I 777N72772O
-------------------------------
272777777777OO
e-----.--------o-e-
2OO 2OO 77O
- e
77O2Oi 77
0 In 0M0NP 772072277N
N2N7N7777 N777777
07327370N7O7
-- N7N772.--
000 0 00 0 3 zI
2--72207777e 77772222e 22e e *
0--*0 0 20 0 00 0
7.. 02- 20-M 02000.-20NNN0 r00m0 OO0 N0 7 - -- 0O 0 ""000000004
2 000000004 7 -
0022222002 03
E- 3
CU 5 I N t H L A T F L UX u I , T k Io U I 140
RUN-PT QW TIN W Po P1 DPlEX XEX OTSUIN DTSU88 Q/WiFG TAU WP/WA DPRAT ZRAT DHR AT DELDH RDER COMMENTS
(W) (F) (LRM/HR) (PSI) (PSI) (PSI) (F) (F) IS)
D7 198 300 114.0 349.0 21.15 20.09 5.26 0.311 20.9 16.2 0.326 2.83 0.417 0.256 0.179 0.756 0.826
08 199 300 103.7 312.0 20.70 19.65 4.91 0.314 29.9 24.2 0.364 3.73 0.491 0.304 0.237 1.402 0.841
011 192 300 94.0 296.0 20.63 19.57 4.91 0.301 39.4 32.8 0.384 4.50 0.54? 0.151 0.3f3 0.779 0.816
D22 98 300 83.6 444.0 20.99 19.86 5.12 0.136 50.6 40.4 0.256 5.60 1.07 0.709 0.523 0.556 0.776 0.690
022 181 300 83.4 275.0 20.37 19.31 4.57 0.293 49.2 41.7 0.413 11.72 2.75 0.620 0.39 0.856 7.711 0.396 7 PEAKS
D22 82 300 83.4 164.5 19.52 18.49 3. 75 0.570 46.8 41.2 0.690 5.10 4.20 0.584 0.284 0.210 0.93 1.110 T5AU CHAN
D23 195 300 69.8 410.0 20.91 19.79 5.06 0.110 64.2 52.4 0.277 6.40 1.15 C.791 0.595 0.360 0.309 0.75?
D23 88 300 71.0 189.0 19.72 18.67 3.94 0.438 59.7 5?.6 0.601 3.00 4.60 0.654 0.353 0.306 0.119 7.065 TAU=6.05
D23 89 300 71.1 166.0 19.64 18.60 3.87 0.522 59.4 52.9 0.684 6.24 4.30 0.629 C.327 0.?70 0.781 1.047
D23 80 300 71.2 159.0 19.56 18.52 3.79 0.552 59.0 52.7 0.714 2.50 4.20 0.627 0.318 0.?78 0.643 2.655
023 81 300 71.3 154.5 19.50 18.46 3.73 0.573 58.7 52.5 0.735 5.49 4.50 0.628 0.313 0.250 0.788 1.231
023 82 300 71.6 142.0 19.36 18.32 3.59 0.639 58.0 52.1 0.800 2.28 3.50 0.675 0.297 0.228 0.765 3.038
D23 83 300 71.7 120.0 19.12 18.09 3.36 0.786 57.7 51.8 0.944 4.02 4.30 0.615 C.268 0.191 0.891 1.839 TAU=2.10
D23 86 300 72.6 84.1 18.32 17.29 2.56 1.193 53.8 49.6 1.350 6.15 7.50 0.643 0.205 ().128 0.663 1.314 TAU=3.07
D21 91 300 42.0 304.0 20.51 19.41 4.85 0.116 90.9 77.9 0.373 6.80 1.24 0.866 0.635 0.716 0.933 1.042 W' TAU?
D21 93 300 42.4 751.0 20.28 19.19 4.63 0.196 89.9 78.7 0.452 7.10 1.19 0.803 0.550 0.597 0.803 1.047 W? TAU?
D21 86 300 43.3 156.0 19.68 18.62 4.06 0.475 87.3 79.0 0.727 7.48 1.80 0.704 0.401 0.373 0.756 1.165
D21 88 300 44.2 141.0 19.36 18.30 3.74 0.555 85.4 77.7 0.805 3.61 5.40 0.712 C.372 0.332 0.719 2.473 TAU=7.12
D21 92 300 42.5 320.0 20.99 19.88 5.32 0.099 91.8 78.4 0.355 6.00 1.09 0.834 0.665 0.758 0.812 1.175 W INCFAS
RUN-PT QW TIN W Po Pl DPlEX XEX DTSUIN DTSUBB Q/WHFG TAU WP/WA DPRAT ORAT DHRAT DELDH ORDER
COMMENTS
(WI IF) (LAM/HR) (PSI) (PSI) (PSI) (F) (F) (S)
D8 191 400 104.2 470.0 22.17 21.05 6.31 0.274 33.4 27.1 0.322 3.29 0.444 0.349 0.301 0.769 0.778
011 193 400 94.6 385.0 21.63 20.54 5.88 0.313 41.6 35.0 0.393 4.50 0.479 0.360 0.316 0.759 0.676
022 82 400 83.5 268.0 20.52 19.46 4.72 0.445 49.6 43.3 0.565 9.72 3.00 0.536 0.327 0.270 0.713 0.412 TAU CHAN
D22 94 400 83.7 420.0 21.66 20.55 5.81 0.241 52.5 44.2 0.361 5.10 1.06 0.562 0.442 0.433 0.779 0.677 W 7
D22 86 400 83.5 283.0 20.94 19.88 5.14 0.415 50.8 44.3 0.535 10.40 2.90 0.516 0.343 0.292 0.705 0.382 TAU=5.08
023 92 400 70.1 430.0 21.47 20.34 5.61 0.187 65.5 55.5 0.352 6.00 1.16 0.657 0.520 0.552 0.772 0.668 W? 1AU'
D21 92 400 41.3 440.0 21.27 20.11 5.55 0.085 93.7 79.6 0.344 6.30 1.05 0.836 0.691 0.794 0.832 0.846
S E R I ES E S T A I L I T Y t X P ck I M t N T b
A L L T E S T L E N 6 T H b P u wQ H t U
RUN-PT OW TIN W PO Pl DPIEX XEX DTSUIN DTSUB8 Q/WHFG TAU WP/WA OPRAT ZRAT DHRAT DELDH ORDER COMMENTS
(W) (F) (LBM/HR) (PSI) (PSI) (PSI) (F) {F) (S)
13 200 61.2 262.0 21.29 20.22 5.39 0.092 74.1 60.2 0.289 0.00 1.09 0.849 0.703 0.723
94 200 61.2 215.0 20.64 19.58
95 20C 61.7 195.0 20.41 19.35
4.75 0.155 72.2 60.7 0.352 3.17 1.21 0.810 0.582 0.599 0.915 2.853
4.52 0.193 71.0 60.6 0.388 3.15 1.23 0.781 0.5?7 0.542 0.826 2.867
16 200 61.8 185.0 20.26 19.21 4.38 0.214 70.5 60.6 0.409 3.10 1.84 0.770 0.500 0.515 0.952 2.914
86 200 62.0 174.0 20.10 19.05 4.22 0.241 69.9 60.5 0.435 2.34 1.60 0.757 C.470 0.483 0.668 3.854 TAU CHAN
17 200 62.0 164.0 19.88 18.83 4.00 0.267 69.2 60.3 0.462 2.31 2.30 0.755 0.442 0.454 0.605 3.894
88 200 62.1 145.0 19.68 18.63 3.80 0.328 68.5 60.7 0.522 5.00 1.24 0.722 0.393 0.404 3.846 1.908 TAU CHAN
19 200 62.0 136.5 19.49 18.45 3.62 0.361 68.0 60.7 0.555 4.98 3.65 0.718 0.370 0.380 0.877 1.815
TES T L ENG T H 1 W I THoUT P Uw E
RUN-PT QW TIN W Po PI DPlEX XEXOTSUIN DTSU8B 0/WHFG TAU WP/WA DPRAT ZRAT DHRAT DELDH ORDER COMMENTS
(W) (F) (LBM/HR) (PSI) (PSI) (PSI) (F) (F) (S)
E04 1 171 78.5 255.5 20.95 19.89 5.07 0.115 55.8 43.3 0.253 0.0C 1.11 0.874 0.638 0.499
E04 92 171 78.6 235.0 20.72 19.67 4.85 0.138 55.1 43.3 0.276 3.50 1.17 0.812 0.598 0.551 0.918 2.396 UNST ?
E04 183 171 78.8 196.0 20.33 19.29 4.47 0.193 53.8 43.4 0.330 3.50 1.23 0.782 0.523 0.461 0.866 0.000 OSC.DIFS
E04 94 171 78.7 185.0 20.17 19.13 4.31 0.212 53.4 43.4 0.350 2.40 1.38 0.780 0.502 0.435 0.741 3.722
E04 5 171 78.8 174.0 20.07 19.03 4.21 0.235 53.0 43.4 0.372 2.52 2.35 0.769 0.480 0.410 0.647 3.608
E04 6 171 78.8 150.5 19.78 18.75 3.93 0.293 52.2 43.4 0.430 2.40 2.50 0.754 0.434 0.354 0.825 3.959
E04 87 171 78.5 130.0 19.58 18.55
E04 8 I71 78.4 120.0 19.45 18.42
3.73
3.60
0.360 51.9 43.9 0.498 7.65 1.61 0.736 0.397 0.30n 0.889 1.314 TAU CHAN
0.401 51.6 43.9 0.540 7.65 4.40 0.731 0.377 0.286 0.909 1.353
TES T L E NG TH S 1 AN 2 WITHUU T P UW LK
RUN-PT QW
(W)
TIN W
(F) (LBM/HR)
PO P1 DPlEX XEX OTSUIN DTSU88 Q/WFG TAU WP/WA DPRAT 7RAT DHRAT DFLDH ORDFP COMMENTS
(PSI) (PSI) (PSI) (F) (F) (S)
1 143 94.5 343.0 21.26 20.19 5.37 0.075 40.7 27.0 0.158 0.00 1.05 0.844 0.698 0.605
92 143 94.3 245.5 20.73 19.69 4.87 0.137 39.4 27.9 0.221 5.0C 1.14 0.796 0.589 0.448 0.92? 1.559 UNST ?
93 143 94.0 215.0 20.40 19.37 4.55 0.167 38.8 27.9 0.252 2.65 1.14 0.797 0.550 0.392 0.785 3.138
4
5
143
143
93.9
93.7
193.0
161.5
20.26
19.93
19.23
18.91
4.41 0.196 38.5 28.1 0.281 2.68 ?.10 0.788 0.524 0.355 0.750 3.295
4.09 0.250 37.7 28.1 0.335 2.70 3.10 0.788 0.484 0.297 0.695
6 143 93.5 110.5 3.608
19.52 18.51 3.69 0.404 36.7 28.3 0.490 2.62 3.20 0.769 0.419 0.?04
87 143 0.825 4.714
93.2 103.5 19.41 18.40 3.58 0.436 36.7 28.4 0.523 6.00 1.76 0.776 0.411 0.192
8 143 93.0 96.5 0.853 2.155 PR 33''
19.31 18.30 9.48 0.473 36.6 28.4 0.561 6.10 6.80 0.781 0.402 0.179
9 143 93.0 148.0 0.836 2.222
19.90 18.88 4.06 0.278 38.3 29.0 0.366 2.75 3.70 0.778 0.472 0.280
10 143 93.0 145.0 0.685 3.775 FIL.BYP.
19.82 18.80 3.98 0.285 38.1 28.8 0.373 2.74 3.45 0.784 0.467 0.273
11 143 92.9 123.0 0.679 3.8325
19.61 18.60 3.78 0.352 37.6 28.8 0.440 2.68 3.40 0.779 0.438 0.231
82 143 92.8 0.795 4.3?8
)6.5 19.43 18.42 3.60 0.472 37.1 29.0 0.561
13 143 92.6 88.3 19.38 18.37 3.55 0.524 37.2 29.2 0.e13
6.06 2.20 0.764 0.404 0.183 0.843 2.252 0-,911.4
6.05 7.25 0.759 0.395 0.168 0.845 2.405
T E S T L E N T H S 11 * 2 AH U 3 W I T H OU T P UW L k
RUN-PT QW TIN W Po P1 DPlEX XEX 1TSUIN
DTSUB Q/WHFG TAU WP/WA DPRAT ZRAT DHRAT DELDH ORUFR COMMENTS
(W) (F) (LBM/HR) (PSI) (PSI) (PSI) (F) (F) IS)
1 114 109.4 420.0 21.23 20.14 5.33 0.072 25.6 12.7 0.103 0.00 1.03 0.812 0.661 0.440
92 114 109.2 364.0 20.97 19.90 5.09 0.087 25.1 12.8 0.119 3.5C 1.06 0.810 0.631 0.386 0.801 1.589 TAU ?
93 114 109.3 ?85.0 20.64 19.60
4 114 109.1 276.0 20.55 19.51
4.79
4.70
0.121
0.125
24.7
24.1
12.8 0.151 3.55 1.14 0.798 0.584 0.301 0.810 1.851
12.8 0.156 3.66 1.80 0.804 0.578 0.291 0.803 1.837
5 114 108.1
6 114 108.8
200.0
102.0
20.28
19.68
19.26
18.68
4.45 0.181 24.4 13.7 0.216 3.75 3.10 0.798 0.542 0.227 0.778 2.321
3.87 0.391 22.0 12.5 0.423 3.38 5.20 0.805 0.475 0.105 0.751 4.422
Appendix F
**~**ut**** ENTHALPY TRAJCCTORY MODEL ******** VERSION 8.2 1/4/70
IMPLICIT COMPLEX (C)
A COMPUTER PRCGRAM FOR THE IBM-360 TO INVESTIGATE THE STABILITY REAL J, CEMM4), CJS, CPROJP, CYCLESCABS
OF A BOILING CHANNEL. REAL MUL, MU. MULF, MUGF
INTEGER*Z NPLOT(60.3501
INTEGER*4 AERASE(105001
INPUT INFCRMATICN DIMENSION 8PL3T(120,'41,ZPLOT(121, XEXK(1201
EQUIVALENCE (NPLCT(1, N'ERASEIX)
FIRST CARC CCMMON// D, OP, GL,ZHEZZMAX, P1 , PRB, PEX, TIN. OTSUBB, DT
CCMMJN /STA20T/ OM(601, 8(6a), E(601, X(6019 10(601, PO(60,
ZPLOT(12) ; COORDINAft ALONG THE CHANNEL IFT) AT WHICH THE TIME I DPDLFRIbOI, DPDLOR(60) DPAC(60), V(60), J(60)
VARIATIONS OF THE PRESSURE ARE TO BE ESTARLISHED; ANY COMMON /OZDEOT/ PUHDw, ADHUW, SDHOW, PDZDW, A02DW, SOOW
NUMBER OF VALUES UP TO 12. 1 PDICWP. AUHUWP, SHOWP A
AA I VALUE OF T1E COEFFICIENT A ICHAPTERS). THE DEFAULT VALUE IS 0.8 2 , PDP10Z, ADP1OZ, POPIDG, ADPIDG# PDPE. ADPE
LCORR 1
IIF'.EO.
USE LINEAR INTERPOLATION TO DETERMINE COMMON/CCMPLX/ CDiCW, 0020W, CUHDWP, COPD12, COPICG, CDP1
PRESSURE AT EXIT; IF .EQ. S CALCULATE PRESSURE DROP IN LAST SEGMENT CCMMON /DYNPR/ 1(47,120),DZDT(40,120), GT40,1201, P(-4C9201
SEPARATELY COMMIN/PLOTOT/ XMAXJ(9), YMAXJ(9), XMINJI91, YMINJ(91, YRANJ(9)
NONLIN : 0-
.EQ. 0 USE LINEAR ESTIMATE CF PRESSURE DROP
IN SINGLE- C
PHASE REGION FROM DZDWTF; IF .EQ. 1
CALL DYNPRI TO DETERMINE BGFIX*ALFAI * (-E.8628E-111*((1.D-XI(R.0-X)1(IS.0-ALFAI*ROL) +
PRESSURE DROP IN SINGLE-PHASE REGION. 1 X*X/(ALFA*ROG))
WHEN THE SAME STEAUY-STATE INFORMATION (lW AND DPIEX) IS SUPPLIED C
FOR TWO SUBSEQUENT POINTS, THE STEADY-STATE CALCULATIONS ARE BYPAS- DATA TWCPI/6.Z83185/
SED FOR THE SECCND DATA WOLD, DPlOLO, TAUCLO /3*0.0/
IREPET -1 PERMITS TO REPEAT THE STEADY-STATE CALCULATIONS EVEN THOUGH O * 0.43/12.0
TAU DID 40T CHANGE AREA - TWOPI * U * D / 8.0
I0MIT -1 PERMITS TO BYPASS THE D2DWTF CALL WHEN
THE SAME PERIOD TAU ZLMAX - 13.0
AND THE SAME STEADY-STATE INFORMATION WAS SUPPLIED BUT CALC- KMAX = 120
ULATICNS START AT A DIFFERENT PHASE, AS SPECIFIED BY TSKIP1 K4AXL - KMAX - I
IWRITE; IF .EQ. 1
PRINTOUT OF THE PRESSURE, MASS FLUX, DZ/DT AND C KMAX MUST BE .GE. 2 AND .Lt. 120
LINES/FT
COORDINATES OF THE MESH POINTS IS PRODUCED. F * 25.0 ~
IF
IPLCTI: .EQ. 1
A PLOT OF THE PRESSURE AT THE BE IS PRODUCOD C F MUST BE SMALLER OR EQUAL THAN 25 LINES/FT FOR NPLOT(XX,350)
IPLOT : IF ,EQ. 1 A MAP OF THE COORDINATES OF THE MESH POINTS C
SIMILAR TO FIGURE 6.3 OFTHE TEXT IS PRODUCED 1 FORMATI12FS.2, F10.2, IX, 711)
2 FORMATITFX0.2. 2X. 2A4)
SELONC CARC 3 FORMAT(1' // 21X, 2A4, 5X, 2A4 /
1 21X,'INPUT INFORMATION TIN -, F7.2, 5X, 'W =s,
TIN i INLET TEMPERATURE (F) 2 FT.1, 5X, 'QWAV -', FT.1, 6X, DPIEX =', F7.2, 5X, *P1
QWAY : GRCSS POWER (WATTS/TEST LENGTH) 3 FT.2, SX, /
PEX s
EXIT PRESSURE IPSIAI 4 T44, 'TAU =', F1.2, SX, *DT(SEC) -', F8.5, 3X, F5.2,
RELAMP I-RELATIVE AMPLITUDE OF THE FLOW OSCILLATION (UW/NOI 5 * CYCLES EXAMINED STARTING AT' F5.2)
YSC : IF .GT. 0 DETERMINES THE SCALE OF THE PLOTSIMAX COORD.) 4 FORMAT400SUBCSB RESULTS 288 -, F8.3, 5X, *P88*
If .LE. ( SUPPRESSES THE PLOTS. IN THIS CASE ONLY THE MAXIMA 1 F8.3, 5X, *DTSUBB = F
F8.21
ARE DETERMINED BY PRTPL3 5 FORMAT('OSTAPR2 RESULIS : IMAX =', 15)
CYCLES : NUMBER OF CYCLES TO BE INVESTIGATED 6 FORMAT('*O,15X,*CM, 1OX,'B*,I0X,'E*,10X,'X,10,X,*Z,1OX,'P'.
TSKIP s DETERMINES THE STARTING POINT OF THE CALCULATIONS; PHASE 1 5X,'DPDLFR*,5X,'UPDL6R*,TX,'OPAC*,OX,V', 10X, 'J' /
OF FLOW PERTURBATION IN CYCLES 2 (15, 5X, 3(IPEI1.31, OPF11.6v 2F11.3. 511PE11.3))l
CCMM(4) I CEMMENTS 7 FORMAT('OCZDWTF RESULTS :*, T20, 'OH (MOD,ARG,DELI :', T40
1 3F2C.5, * BTU/18M' /
THIRD AND FOLLOWING LARUS 2 T20. 'DZ1, T40, 3F20.5, * FT' /
3 T20, 'HP', T40, 3120.5.RTU/LBM* /I
W I AVERAGE MASS
FLOW RATE ILBM/HRI 4 T20. 'M102', T40, 2F20.5. 20X, ' PSI' /
OP1EX : PRESSURE DROP FROM STATION 1
TO EXIT (PSII 5 T20, 'PIDG, T40, 2F20.5, 20X * PSI* /
TAU : PERIOD OF THE CSCILLATION 4SI 6 T20, *PI ', T40, 2F20.5,20X,* PSI*,10X,'(A =,F5.2, *l' I
TSRIPX 3 SAME AS ESKIP BUT FOR SUBSEQUENT POINTS. PERMITS TO BREAK 8 FORMATIOCDGPEAK =', F1C.1, * LSM/HR-FT2*, 5X, 'DZPEAK -* F8.5,
lHE ENTIRE CYCLE INTO MORE THAN ONE CALCULATIONS (SPECEFY 1 ' FG*, 5X, 'UUZDTP -'. FE0.5, ' FT/SEC' I
ADEQUATE TSKIP, TSKIPI AND CYCLES) 9 FORMAT(*0DYNPRI : Ptl,Kl 1* / 410F13.61)
PICOR : CORRECTION FACTOR TO MAXIMUM AMPLITUDE DF POSITION CF THE 10 FORMAT('1', 5X, 3014 / )
BB PERTURBATION. DEFAULT VALUE IS 1.0 11 FORMAT('ODYNPR2 : P1,K1 (PSIAI' /
ADZCOR : CORREC TION (IN CYCLESI TO DELAY OF POSITION OF BB PERTUR- 1 BX, 2016 / (15, 5X, 20F6.2))
BATION. DEFAULT VALUE'IS 0.0 - 12 FORMAT(*1DYNPR2 :Z{I,K lFT * /
PEXCUR : VALUE
OF THt EXIT PRESSURE TO BE USED WHEN THE PRESSURE DROP I 8X, 2016 / (15, 5X 20F6.311
IN THE LAST StGMENT IS CALCULATED SEPARATELY. DEFAULT VALUE 13 FORMAT('0', 5X, 3014)
IS EQUAL TO PEX. CAN RE USED TO SIMULATE THERMAL NON-EQUIL- 14 FCRMAT(IOOYNPR2 :G,K) 10**5(LRM/HR-FT2)' /
IBRILUM EFFECTS. 1 1X, 2016 / (15, 5X, 20(-5PF6.31))
15 FORMAT('OCZOWTF : BOILING SOUNDARY IS ON A COLD SPOT 'I
16 FORMAT('ODYNPR2 : I2DT1,IK) FT/HRI' /
I RX. 10112 / {15, 5X, 10F12.11I
IT FORMAT)'OREL. AMPLITUuE =', F6.3) C
18 FORMAT('0********-* FOR A RtLATIVE AMPLITUDE OF', F5.2, C DETERPINE PCSITIDN OF BOILING BDUNDARY . . . . . . . . . . . . . .
1 9 THE RkLATIVE CENTLR OF VOLIUMt VtLOCITY AT THE BOILING', C
2 / 12X,'BOUNDARY MIGHT BECOME NEGATIVE', .............. CALL SUaCEB
3 'CALCULATI:N DISCONTINUED') C
19 FORMAT('OROILING TRANSIT TIME ='. F8.3, ' SEC') C
20 FORMATI(IS, 3X, 6012)) UTSEC - TAU 0 CYCLtS / FLC4TIKMAX)
21 FORMAT)'l' // 113 OT - OTSEC/1600.0
I 'OSAPE STEAnY STATE INFORMATION AND SAME DT USED NOW FOR C
2 'A CIFFERENT PE-fOD OR TSKIP:'/'OTAU *, F6.2) WRITE(6,3) CUMM, TIN, W,
QWAV, DPIEX, PI, TAU, DTSEC, CYCLESTSKIP
22 F-ORMAT(7F10.2, 2X, 2A4) WRITE46,L7) RELAMP
23 FORMAT(15, 38X, F11.4,
2F11.3, 33X, 2(I1PE11I.1))- WRITE16,4) 18E, P8, DTSU8D
24 FORMAT)'0.........', 4A4, '.........'/) C
25 FORMATf'OPktSSURE AT TMt EXIT FROM TIME', FT.4, * CYCLES TO TIME', C DETERMINE REFERENCE STEADY-STATE CONDITIONS IN THE CHANNEL . . . .
1 F7.4, ' CYCLtS'/ (5X, 15F8.4)) C -
26 FORMAT//'0******** OT(StC INCREASED BY A FACTOR OF
1.3 IN , CALL STAPR2(IMAX)
I 'OGRDER TO REDUCE NUMBER OF MESH POINTS *********'//) C
27 FORMAT(*OPEAK OF LXIT PRESSURE OUTSIDE RANGE, AT K
*', 13) IF(IMAX .EQ. 1000) GO TO 114
28 FURMAT(*0*********** 10(', 12. ')SMALLER THAN 9.30207, IMAXI - IMAX - I
I CALCULATICN DISCUNTINUED *=*********'//) WRITE(6,5) IMAX
29 FORMAT)'0', ' REL. AMPLITUDE ' , F6.3, 20X, F5.2, ' CYCLES EXAMI WRITE(6,6) (I, IM(II,t(I),E(i),X(I),Lt(I),PU(I)DPOLFR)I),
1NED NOW STARTING AT', F5.2) 1 DPOLGR(I),DPAClI),V(I), J(I, I=1,1MAX1)
30 FORMAT('CbXIT PRESSURE, PEX ' , F7.3, 10X, 'PRESSURE DROP IN TWO-' WRITE(6,23) IMAX,X(IMAX), ZO(IMAX), PO(014AX), V(IMAX), J(IMAX)
1 , 'PHASE REGION, PBS-PEX -', F7.3 / 'OPEAK TO PEAK VARIATI', C
2 'CNS : PEX :', FD.4, SX, 'TWO-PHASE PRESSURF DROP tI, C CALCULATE THE TRAASIT TIME IN THE BOILING REGION . . . . . . . . .
3 F1C.4, ' PSI', SX, 'EXIT FLOW :', F10.0, ' LBM/HR-FT2' 1 C
31 FCMAT('O************ RDVEMENT OF 80LING BOUNDARY CORRECTED', 00 49 1 - lIMAX
l ' EXTERNALLY :') IFIZO(I) .GE. 9.30207) GO TO 48
32 FORMAT(IX, 2A4, F6.1, F6.2, 2X, 6(F6.3, F5.3, 1X), lX, -5PF6.3, 49 CONTINUE
1 OPFS.3, 1X,
F6.3, 2X, F6.3, F5.3) SUMMARY WRITE(6,20) IMAX
33 FORMAT(/'OS U M M A R Y 0 F R E S U L T S , THRESHOLD POINTI', GO TO 1111
1 2A4, ' AT', F7.2, ' WATT/TS'// 48 BOLTRT = L(I-1)*DT+ALOG(I.+ODM(I-1)*E(l-1)*(9.30207-Z1-1))
2 2X, 'POINT', T13, 'i', T18, 'TAU', T26, 'DMOW', T39, 'DZDW', 1 /(L0(I)-L2(I-I))) / JM(I-1)) * '600.0
3 T52, 'DP1DZ', T64, 'DPlDG' , T76, 'DP1', T85, 'DP2(ENTH)', WRITE(6,19) BOLTAT
4 T99, 'DG(FBK)', T110, 'RATID', T120, 'OPEX' /) IF(IMAX-40) 112,112,115
34 FORMAT('0EXIT QUALITY : / (5X, 20F6.3)) 114 DTSEC = CTSEC*1.3
35 FORMAT('CEXIT PRESSURE DETERMINED RY LINEAR INTERPOLATION') 115 TSEC - ISEC*1.3
36 FORMATI'OEXIT PRESSURE DETERMINED BY CALCULATION OF PRESSURE DROP CYCLES - KMAX*CTSEC/TAU
IIN LAST SEGMENT'/ IF(IOMIT *LE. 0
2 ' IN ALL CASES SEGMENT STARTED BEFORE ZCORR *', F5.2, 1TSKIP - 0.5*(1.0-CYCLES) - 0.5*SDLTRT/TAU
3 FT ADJUSTED EXIT PRESSSURE, PEXCOR = ', F6.2, ' PSI WRITE16,26)
4A') GO TO 113
C C
CALL ERASEIZ,4800, D0OT,4800, G,4800, P,4800) C
C 112 IF((ICMIT.GI.)0.AND.(TAU.EU.TAUOLO).AND.(W.EQ.WOLD).AND.(DP1EX.EQ.
C REAC AND PROCESS INPUT INFORMATION . . . . . . . . . . . . . . I OPIOLD)) 60 TO 116
C C
READ(5,2) ZCCRR C CALCULATE PEATUReATION tF THE POSITION OF THE RB AND SINGLE-PHASE
1110 READ)5,1,END-9999) ZPLOT, AA. C REGION DYNAMICS . . . . . . . . . . . . . . . . . . . . . . . . .
1 LCORR, NONLIN, 10DIT,IREPET, IWBITE, C
2 IPLCT1, IPLOT CALL DEbTFTAUICOLD)
READ)52,END*9999) TIN,QWAVPXRELAMP,YSC,CYCLESTSRIP, C
1 CCPM(1),COMp(2) IF(ICCLU .EQ. 1) WRITE(6,15)
IF(TIN .LE. 0.0) STOP WRITE46,7) POMOW, ADHDW, SDHDW, PDz1W, AOZuW, SDZDW,
A - 0.8 1 PDFCWP, ADHUWP, SOHOWP, POP1DZ, ADP1UZ, POPIG, AOPlOG,
IFAA .NE. 0.0) A - AA 2 POPE, ACP1, A
WRITEI8,33) COAMI), COM12), QWAV C -
1111 RCAO5,22,ENDO9999F W,
0PIX, TAU, TSKIPI, POZCO, ADZCOR C
I PEXCOR, COMM(3), COMMI4) C CORRECT IF NECESSARY THE AMPLITUDE OR THE PHASE OF THE POSITION OF OR
IF(PEXCOR
.LE. 0.0) PEXCOR * PEX C PERTURBATIOn . . . . . . . . . . . . . . . . . . . . . . . . . . .
IF(TAU .LE. 0.0) GO TO 1110 C
IF(IOMIT .LE. 0) GO TO 117 116 IF(POZCCR .EQ. 0.0) PCZCOR = 1.0
TSKIP - TSKIPI IF(IPDCOR .EQ. 1.0) .AND.IADZCOR .EQ. 0.0)) GO TO 118
117 I1MW .NE. WOLD) .0R. IOPlEX .NE. DPIOLD) .OR. IIREPET .EQ. 1)) ROL = TWCPI*ACLCOR
I GO TO 111' CDZCOR v CMPLX(COS(ND),SIN(RUZIb*PPLZCUR
WRITE(6,21) TAU CDZOW = CCZCW*CCZCOR
WRITE16,29) RELAMP, CYCLES, TSKIP COPIOZ = CCPIDZ*COlCOR
GO TO 112 CDP& - CCPI0Z * COPIDG
C C
111 P - PEX + DPlEX PDLDW CAOSCDZDW)
=
QP = (3.412/1.291667)OQWAV ADZOW AMO0(1.0-ATAN2AIMAG(CDZDW),REALCDUZDW))/TWCPI),1.O)
=
QTP - QP/AREA SOZDw TAU * ADZDW
=
G0 = W/AREA PDPIDZ - CABS(CDPID)
ADP1OZ = AMUDI(1.O-ATAN2)AIMAG(CDP1DZ),REAL(CDPI0Z))/TWOPI),1.0) (CYCLES)
PQPl - CAS(CDP1 )I I(PSI WRITE(6,14) (Li L -1,K20,KDK),
ADP * AMO((l1.0-ATAN2(AIlAG(COP I,REAL(COPI ))/TWOPI),1.0) (CYCLES) 1 It, IG(II,K), K-,K20,KDKI, 1.1,IMAX)
WRITE(6,31) -WRITE(6,16 (L, L =1,920,KDKI,
WRITE(6,1I PUHDW, ADHa, SDHDW, PUIOW, ADZDW, SOZUW, 1 (I, (ZDT(IK), K-1,K10,KDK2I, I-1,IMAXI
S PDIDWP, ADHDWP, SHUdP, POP1DZ, ADP10Z, POPIG, ADPE0G, WRITE16,24) COMM
2 POP1, ADP1, A C
C 187 IMIPLOT .LE. 0) GG TO 410
C ZMAX = -1.CE50
C GENFRATE C(1,K), ZI1,K AND ODZOT(I1,K) . . . . . . . . . . . . . . ZMIN = 1.0E50
C C
118 OGPEAK - RELAmP*GC DO 199 K =I,KMAX2
DZPEAK = RELAMP* PDZOW 'IFIZMAX-Z(IMAX,K) 191,191,199
DOZOTP - 0ZPEAK*T4OPI*3600.0/TAU 191 ZMAX - J(IPAX,#K
IFI(GO-OCPEAK) * V(I),DUDTP) 50,50,51 199 CONTINUE
50 WRITEI6,10) RELAMP C
GO TO 1til DO 299 K -1,KMAX1
51 T - TWCPI * ADZDW IFIZlEE) - ZMIN) 291,291,299
DTOIN = CTSEC/TAU 291 ZMIN - Ztl,K)
UTIME * 7tAPI*DTUIM 299 CONTINUE
TINE = -OTIME + tSKIPTTWOPI C
C C PRODUCE A MAP OF THE COORDINATES Of THE MESH POINTS . . . . . . .
DU'9 K *l,,jMAX C
TIME TEME + OTIME CALL ERASEINERASE,105001
G(1,Kl GO + DGPEAK*CCS(TIME) C ERASE IS A CENTER SUPPLIED SUBROUTINE FOR ZEROING LENGTH *4 ARRAYS
141,K) - Z88 + DLPEAK*COS(TIME - T) C
99 D20T44,KI -DDZTP*SIN(TI4E-T) KOK kNAX1/60 + 1
C KDK2 2*KDK
OUZOTP = DDZDTP/3600.0 IF(ZMAX - 13.96) 351,352,352
WRITE(6,81 DGPEAK, CZPEAK, ODDOTP C
C 352 D0 389 1 = 1,IMAX
C KKK - KMAX - I
C IF A NONLINEAR TREATMENT OF THE SINGLE-PHASE REGION IS SPECIFIEW' 00 389 K - 1,KKkKOK
C DYNPRI IS CALLEO, OTHERWISE THE DZDVTF RESULTS ARE USED IFIZlIK) .GE. 13.961 GO TO 389
C TO CALCULATE THE PRESSURE DROP IN THE SINGLE-PHASE REGION NPLCT(KZ(IK)*F1I) - I + 10
C 389 CONTINUE
IF(NONLINI 60,60,61 GO TO 400
C C
60 TPI * TWCPI * AGP1 351 00 399 I * 1,IMAX
TIME = - CTIME + TSKIP * TWOPI KKK - KMAX - I
00 43 K 1,KMAXI 00 399 K -1,KKK,KDK
TINE * TIME + CTIME 399 NPLOT(K,Z4I,Kl*F+1I) = I + 10
63 P(1,K) - PB + RELAMP*PJPI*COSITIME-TP1) C
GO TO 62 40C IMIMIN - ZMINMF - 1
C IMAMAX = ZMAX*F + 1
61 CALL OYNPRIIKMAX) IFIZMAX .GT. 13.96) IMAMAX = IFIX(13.96*F)
C WRITE(6,104 (I, I =1,KMAX,KD3K2l
62 IF(IWRITE .GT. 0) WRITE(6,9I (P(I,K), K = 1,KMAX WRITE(6,20) tI, (NPLOT(K,II, K -1,601, 1 = IMIMIN,IMAPAX)
C WRITE(6,13) (I, I =1,KMAX,KDK2)
C PLOT THE PRESSURE AT THE ROILING BOUNDARY VERSUS TIME. WRITE(6,241 CCMM
C C
YSC1 = 1.0 C FIND TIME VARIATIUNS OF PRESSURE AT GIVEN LOCATIONS ZPLOT(Il
IFIIPLOTI .LE. 0) YSCI - -1.0 C FIND MIN VALUE OF ZIIMAX,Kl
C C
00 109 K * 1,KMAX1 410 ZMIN * 1.CE+50
BPLOT(K,1) - (K - 1I*OTDIM + TSKIP D0 479 K =1,KMAX2
109 RPLDTIK,2l * P(I,K) IFtZIIMAXKI-ZMIN) 47T1,471,479
CALL PRTPL3I1,RPLOT, KMAXI,2, KMAXI,0, KMAX,4, YSCII 471 ZMIN * ZIIMAX,K)
C 479 CONTINUE
C USE THE ENTHALPY TRAJECTORY MODEL FOR THE DYNAMICS OF THE TWC-PHASE C
C REGION . . , . . . . . . . . . . . . . . . . . . . . . . . . . * IF(ZPLOTtll .LE. C.0) GO TO 500
C C
CALL GYNPR2(IMAX, KMAXKMAX2) C FINC MAX VALUE OF Z(1,K1
C C
KMAX3 a EPAX2
- 1 ZPAX - -E.OE+5,0
C 00 489 K -1,KMAXZ
IFI(WRITE ,LE. 01 GO TO 187 IFIZMAX - AI1,RI1481,489,489
C 481 ZMAX k ZI,K
KOK * KPAX1/20 + 1 489 CONTINUE
KDK2 - 2*KDK C -
K20 * 19*KOK * 1 00 499 L =1,12
10 - 9*KCKZ + 1 ZP - ZPLCTILI
WRITEI6,12 IL, L -1,K20,KDK), IF(IP .LE. 0,0) GC TC 500
1 (I, (ZII,K), K-1,K20,KDKI, I=leIMAX) IF(ZP *GT. 2MIN) GO TO 499
WRITEt6,111 IL, L =1,K20,KDK)i, IF(ZP .LE. ZMAX) GO TO 499
1 (I, (PII,9), K-l,K20,KDK), I-I, IMAXI
M =0 C DETERMINE EXIT WUALITY ACCURDING TO PEXCOR, CORRECTED EXIt
00 498 K - i,KMAX3 C PRESSURE 'TO TAKtt INTO ACCOUNT Sflt THERMAL NON-EQUILIRRIUM
M M + I C
DO 49T 1 - 1, IAX WRITE(6,361 ZCCRR, PtXCOR
IF(ILK) - ZP) 497,496,496 HFGEX = bfGF(PEXCOR)
496 IF#Zil-1,K+11 .GT. ZP) GU TC 494 HEX - HLFIPEXCOR)
C XII = X11)
C DETERMINEPRISSUE AT ZP BY LINEAR INTERPOLATION HII = HLF(POIIIII + XII *HFGFIPO(II)
C X1 = EHII - FEX)/HFGEX
BPLOT(1M,21PII-1,K+I)+(P(I,K)-P( I-1,K+1))*(ZP-Z(I-1,K+1))/(Z(1,K) C
I - Z(I-1,K+1)F C ESTIMATE PROPERTIES AT AVERAGE PRESSSURE
BPLOT(M,1) - (K + I
- 21 * uTDIm + TSKIP
C
GO TO 498 PAV - 0.5*IPO(II) + PtXCOR)
497 CONTINUE MUL * MULF(PAV)
494 M- M - 1 MUG - PUGF(PAV)
498 CONTINUE ROL = RCLFIPAVI
CALL PRTPL3(IFIX(100.*ZP+.5),BPLOT, M,2, M,0, KMAX,4, YSC) ROG - ROGF(PAVI
499 CONTINUE ROGROL * ROG - ROL
500 CONTINUE XTTI - (ROG/R(TL)**0.5 - (MUL/MUG)**O.l
C XTTII * XIII *tE,0-XIII/XIII**0.9
WOLD =w ALFAII - ALFAFtXTTII)
UPIOLD - OP1EX BGII - 8GF(XIR, ALFAIII
TAUOLD - TAU C
C C ADD ENTIALPY VARIATIGNS AT ACTUAL POSITION OF R3 TO ENTHALPY AT
C FIND FLEW AND PRESSURE VARIATIONS AT EXIT AND UP IN TWO-PHASE REGION C MESH POINT If
C C
ZP - 9.30207
PRBEX- PBB-PEX
M - KMAX2 + 1MAX - II - I
DU 559 K - 1,M
C, GIl - G(I,K)
IF(LCORR .EQ. O1 GO TO 550 D1 - ZP - Z(II#K)
IF(ZP .GT. ZMIN) GO TO 550 TIME = TWOPI*TSKIP + (K-1)*JTIME
C XEX a Xi + IQTP*I(0-0.26)/GII + RELAMP*PDHDWP*COS(TIME-ADHDWP*
C USE LINEAR INTERPOLATION TO DETERMINE EXIT PRESSURE. 1 TWOPI)P/HFGEX'
C XEXKIK) - XEX
WRITEI6,351 XAV - 0.50(Xl1 + XEX)
M 0
XTTtX = XTTi*tEt.0-XEXI/XEXI**0.9
DO 548 K = I,KPAX3 XTTAV * C.5*IXTTII + XTTEX)
M =M + ALFAAV * ALFAF(XTTAV)
DO 547 1 - 1,1MAX ALFAEX = ALFAFIXTTEX I
IFIZ(I,K) - ZP) 547,546,546 XUMUL- (I.0-XAV)*D/MUL
546 IF (Z(I-I,K+1I .GT. ZPI GO TO 544 OPFRIC - FI2LFIXTTAV)*(-3.3256E-11)*IDE/DI*FF(GII*XDMUL)*GI **2
CPROP = IZP-Z(I-iK+1))/ (Z(IK)-Z(I-I,K+1I)) I *tl.0-XAV)**2 / ROL
BPLOTIM,I) = (K + 1 -2) *DTUIM + TSKIP DPGRAV = -CL*IROGROL * ALFAAV + ROL3/144.0
BPLOT(Mt2J - IP(I-i,K+i) + CPRUP * (P(I,KI-P4I-1,K+1)I)/PEX DPACCE = (OGFIXEX, ALFAEX) - RGII)*GII*GII
BPLOT(M,3) = (G(I-1,K+1I + CPROP*(G(I,K) - G(I-1,K+1)))/GO dPLOTIK,1) - (K+I-21*DTDIM + TSKIP
BPLOT(M,41 IP(I1,K+1-1)-BPLOT(M,21*PEX)/IPBEX) 8PLCT(K,2) - fP(II,K)+ OPFRIC + DPGRAV + DPACCEI/PEX
GO TO 548 BPLOT(K,31 = GII/GO
547 CONTINUE 559 9PLOT(K.4) - PiIK11-1 - BPLOTIK,2)*PEX)/PR8EX
544 M = P - 1
C
548 CONTINUF CALL PRTPL3(930,BPLOT, M,4, M,0, KMAX,4, YSC)
CALL PRTPL3(93C,RPLOT, ,4, M,0, KMAX,4, YSCI WRITE(6,24) CCMM
WRITE16,24) CCMM PEXRAA = YRANJ(I) *PEX
C WRITE46,25) BPLCT(I,1), BPLOTIM,I), (BPLOTIK,2), K =1,M) GEXRAN = YRANJI2) * G3
PEXRAN = VRANJIII *PEX DP2RAN = YPAAJ(3)
* POBEX
GEXRAN = YPANJ121 * GU WRITEIE,3C) PEX, PBREX, PEXRAN, DP2RAN, GEXRAN
DP2RAN = YRAAJ(3) * PROEX WRITE16,34)(XEXK(K),K - 1,M)
WRITE6,301 PEX, PBSEX, PEXRAN, UP2RAN, GEXRAN C
C CONTINUE
600
C C
550 IFILCORR .GT. 0) GO TO 600
C
C C CALCULATE OPEN LOCP TRANSFER FUNCTION
C EVALUATE PRESSURE DROP IN LAST SEGMENT INDEPENDENTLY C
C FIND ZMIN, ZMAX,
FUR IM4AX, IMAX-1, ... UNTIL ZMAX .LE. CHANNEL AVP2E * XPAXJIJl
C LENGTH. PDP2E - 0.5*P?RAN / RELAMP
C C
II = IMAX + 1 ROP2 = TWOPI*ADP2E
551 11 = I1- 1 COP2E = CPPLX(COS(-RDP2ISIN(-ROP2I)*PDP2E
ZMAX = 0.0 CDPIt -- CCP2E
00 552 K = 1,KMAX2 COPIG - CDP1E - CUPlCZ
IFtZfII,K) .GT. ZMAX) 14AX - Z111,K) COGFDR =-CDPiG*GO / CUPOG
552 CONTINUE PDGFEB = CABSICDGFU8)
IF((ZMAX.GT.ZCORR).AND.(II.GT.1)) GO TO 551L ADOGFDb A3OD((I.0-ATAN2(AIMAGICL)GFDB),REAL(CDGF09)/ITWOPI),I.0)
IFIZMAX .Gt. ZP) GO TC 600 RATIC = PCGFDR / GO
WRITE(8,321 CCMM(3), C0M(4), W, TAU, POHLIW, 4M4OW, PDZDW, A0ZOW,
1 PDPIDZ, APICZE, PJPiG, AOPI0G, PDPI, ADPI, PDP2E, ADP2E,
2 PDCFC , ADGFOB, RATIO, PEXRAN, XMAXJ(I) SUMMARY
GO TO 1111
9999 STOP
END
P(1+1) = PI1) + DPFR + JPGR
PSATII+I) = PSATFit(Il+1))
SUBROUTINE SUBCBB Z(1+1) = 211) + DL
C VERSION 3.1, 11/3/69 IfT-(Pl+1) .LE. PS&tIl+1)) Go TO 2
C TAKES INTO ACCOUNT C-LD SPOTS IFIl *Lt.15) GO TO I
C DTSU8 = t(l6) - TIN
C CALCULATE SURCOOLING WITH RESPECT TO BOILING BOUNDARY FDR EXPERIMENTAL P88 * P116)
C PClNTSiSTARTING PRESSURE DROP CALCULATIONS FROM
INLET. 288 - 9.C41668
C INPUT I CP IBTU/HR-FTh# G (LBM/HR-FT2), PIN - Pi (PSIA), TIN (F) WRITF(6,9)
C OUTPUT : 288 (FT), PB6 IPSIA), DTSUBB IF) 9 FORMAt(00******** SUbC B : BOILING aCUNDARY IS OUTSIDE HEATEDN
C FUNCTICNS REUIRED I
RULTF, MULTF,
HLTF, PSATF, TF, FF KLF I t* SECTION' //)
C RETURN
REAL PULB6MULw,MULtF, KLF
REAL QDISTRIP)/?*T.0/ FIND EXACT LCCATICN OF BOILING BCUNDARY BY ITERATION . . .. . . .
DIMENSIEN DISTR(15, 21(15),
T1161, P(L6)o PSAT(16), H(16)
COPRON // Do 6P, G, Z8, LMAX, PIN, P88o PEX, TIN, DTSUBB, DT 2 DLP - OL
DATA DLH+ DLU /1.125, 61666667/ 6 DLP - DLP * IPWII - PSATII)) / (P(I)-P(I+I1)+PSA(1+1)-PSAT(I))
C H(I1+1) = DH * OLP/DL + H(I)
C ********* FRECN - 113 PROPERTY FUNCTIONS ****** it1) TF(HIl+1))
C TAV - 0.5 * (TI)) + t1+1))
TIT) 1.0 / It + 459&61 ROL - RCLTF(TAVI
ROLTFITI 4 .D305453 E 03 - .7105255 E-01*T - .6448694E-04*t*T ROLTFI/1 PULB = PLLTF(TAV)
PULTF(T}* EXP( - .3483695 E 01 + .5585921 E 04*TI(T) TWAV - TAV + DTFILM
I - .1950534 E CT*TI(T)**2 + .3577018 E 09*TI(TI**3) MULTF PULW
= PULTF(TWAVI
HItFiT)- *.119482E GE + .1961794E 00*T 4 .126I04tF-03*T*T HLTF 1/1 REL G * 0 / PUL3
6
PSAtF(T) I.0**133.C655-4330.98/(T+459.6) - 9.2635*ALCGIO(t+ DPFR w (-3.3256E-11)*(DLP/)I*FF(tL)*G*G*IIMULW/UL8I**0.14)/ROL
1 459.6 + 0.0020539 * (T+459.6)1 PSATF OPGR = -OLP *ROL /144.0
TFIH) - .4C8?034E 02 4 .51866R1E 01*H - .1141203E-O*H*H TF 1/1 P(1+1 a PI T) DPI- +-DPGR
RLF(T) * C.0475 - 6.?666TE-05* T KLF 1/1 PSAt(I+ - PSATFIT(1+1))
C IFIABS(P(I+1)-PSAT(I+1)) .GT. 0.01) GO TO 6
W *OS165A*O*D*G DTSUBB -k 1T+1 - TIN
C P88 P141)
C IENERATE CORRECTED FEAT FLUX DISTRIBUTION.. . . . . .. . Z88 = 2) + DLP
C RETURN
RATIO * IU00 + DLH)/ULH END
DC 98 J 2,14,2
98 DISTR(J) GQISTRJ/2) * RATIO
C
C GIVE INITIAL VALUES At HEATED SECTION INLET . . .. . .. .. .
C SUBRCUTINE CZOWTF(TAU, ICOL)
til) = 0.0 VERSION 3.2, 12/0/69
Hit) = HLTF(TIN)
TI() - TINR SUBROUTINE OZOWTF CALCULATES THE AMPLITUDE AND PHASE OF THE
Pit) -PIN BOILING BCUNCARY USCILLATION USING A LINEAR MODEL.
PSAT11) - PSATFWIN)
C IN THIS VERSICN THE CCLO SPnTS IN THE WALLANU THE HEAT FLUX DISTR-
C CALCULATE BULK TEMP. AND SATURATION TEMP. AT THE END Of EACH INTERVAL IBUTICN ARE TAKEN INTO ACCOUNT
C THE PRESSURE CROP VARIATIONS IN THE SINGLE-PHASE REGION ARE ALSO
CALCULATED.
I 0 +
1 1 =1.+1
If-IMCCII,)I 3,304 INLET VARIABLES
3 UL = DLH TIN t INLET TEMPERATURE IF)
OH CL*QP*DISTR (I) / W DTSus84 SUBECCULING WITH RESPECT TO THE BOILING BOUNDARYSAT. TEMP IF)
H(I+1) * i-I) + OH PL: PRESSLRE AT INLET CF HEATED LENGTH (PSA)
Tili+1) = YTPINI+1-) QP 1 AVERAGE LINEAR HEAT RATE (BTU/HR-FT)
TAV = 0.5 * (TII) + T(I+1)) G AVERAGE MASS FLUX (LBM/HR-FT2I
ROL = ROCLTFITAV) 2881 DISTANCE OF BOILING BOUNDARY FROM INLET (FT)
MUL - MULTFITAV) TAU t
PERICO Of SINUSOIDAL OSCILLATION (SEC)
OTFILM = 1282.*D.L)*QP*OISTR(I)*MULB**0.4/(G**0.8*KLF(TAV)**0.6)
tAAV- TAV + OtFILM QDISTRIt) DIMENSIONLESS POWtR DISTRIBUTION (INLET TO EXIT)
MULW * AULTFITWAV) A : COEFFICItNT DEFINED BY 0Q/Q = A * OW/W
GO TC
5
4 0L - DLU1 OUTPUT t
IFI .EO. IA UL * 0.5*DLU DHDW 1 INLET FLOW TO ENTHALPY AT STEADY STATE LOCATION OF 88
DH a 0.0 (UITH WALL
EFFECT), IBTU/LBM)
H(11) & H() ZOW : INLET FLOW TO POSITION OF ROILING BOUNDARY (ALL EFFECTS
TAV TIE) INCLUDEC) (FT)
T uil1TAV DHDWP : INLET
FLCW TO ENTHALPY AT MOVING POSITION Of RE, (STU/LBM)
ROL * RTLTF(TAV) DPIUZ : MOVEMtNT OF THE B4 TO PRESSURE DROP IN THE SINGLE-PHASE
AULB- PULTFIAV) REGION (PSI), FRICTICN AND GRAVITY EFFECTS.
OTFILM * 0.0 OP1OG : INLET FLOW TO PRESSURE DROP IN THE SINGLE-PHASE REGICN
MULW VLLS
P (PSI), FRICTION ANC INERTIA EFFECTS
5 REL =G * D MULB- OP : INLET FLCW TO TOTAL PRESSURE DROP IN THE SINGLE-PHASE REGION IPSI)
OPFR = (=3.3256E-1t)*(CL/0)*PF(REL)*G*G*(IMULW/MULBI*=0.141/ROL ICOLD : EQ. 0
(F THE BB IS IN THE HEATED ScGMENT, .t . I IF THE 88
nOGR = - T" * ROL / 144.0 IS IN THE COLD WALL
C
C THE PREFIX C DENOTES THE COMPLEX JUANTITY 881 NTS - I
C THE PREFIX P GENCTES THE MAGNITUDE (PEAK VALUE) GU TO 89
C THE PREFIX A DENVTES THE PHASE (IN CYCLES) 882 111+2) - L(1+1) + DZU
C THE PREFIX S DENOTES THE DELAY (SEC) IF(ZAIB-Z(1+2)) 488,q83,88
C 883 NTS - I + 1
IMPLICIT CCPPLtX (C) GC TC 89
REAL PULTF, KW* CAdS, KLF, MULB 88 CONTINUE
DIMENSICA QDISTR(7),
COEXP(16), DISTR(15), Z(16) 81 NTS = NTS + I
COMMON// 0, .JPG, LU6,
LMAX, Pl, P88, PEX, TIN, DTSUBB, OT Z(NTSXI - 2P3
COMMON /CLOWOT/ POHDW, ADHDW, SDHW, PDZOW, ADZDW. SDZDW. [COLO = MCD(NTS.2)
I PDHDWP, AOHDWP, $DHDWP, A 00 87 I = 2,NTS1
2 , POP10Z, AUPECZ, PDP1UG, ADPIG, PDPE, AOP1 87 COEXP(I) - CEXP(CK*ZIll)
COMMOA /CCMPLX/ CUHCW, CDZ.., CDHOWP, CDPEOZ CDPE0G, COPE C
DATA DISTR/15*0.0/, WUSTR /7*1.0/ ******* C EVALUATE INTEGRAL
DATA PITWOPI/3.141593, 6.283185/, CI/(0.0, 1.0)/, C
I ALPHA, KW, THICK /0.022, 0.75, C.004/ M.I.T. CSU4 =0.0
2 , DlH, OL/1.125, .1666667/ DO 86 I
= 2.ATS,2
C
M.I.T. 86 CSUM * CSUM + DISTR(I)*(CEXP(I+1) - CJEXP(II)
C PROPERTY FUNCTICNS RE;UIRE: ROLTF, MULT, DHLF, KLF, HLTF C
C C CALCULATE THE INLET FLCW T9 ENTHALPY TRANSFER FUNCTION...
TI(T) - 1.0 / (IT * 459.6) C
MULTF(T)= EXP( - .5483695 E 01 + .5585921 E 04*r[(T) MULTF1/2 CGI = CL/CK
1 - .1950534 E 07*TI(T)**2 + .3577018 E 09*TI(T)**3) MULTF2/2 CG2 = -CSLM/COLXPANTSI)
ROLTF(T) - .1035453 E 03 - .7105255 E-01*T - .6448694E-04*T*T ROLTFE/E CCHDW = CGI * CG2 (STU/LBI
HLTF(T) = .81194F2E 01 + .1961794E 00*T + .1261047F-03*T*T HLTF 1/1 C
KLF(Tl = 0.0475 - 6.76667E-05* T KLF 1/1 PDHDW = CABS(CDHDW)
C ADHOW = -AT8N-2IAIMAG(CCHDW),REAL(COHDWI)/TWOPI
C AVERAGE TEMPERATURE AND OENSITY IN NON BOILING REGION IF(ADHDW .LT. 0.01 ADHOw = 1.0 + ADHDW
C SDHDW = TAU * AHW)W
TAV * TIA + 0.5*OTSUfd C
ROLAV = ROLTFITAV) C CALCULATE THE INLFT FLCW TO POSITION OF THE 8.8. TRANSFER FUNCTION
ROLIN = RCLTFITIN) C
MULB = sULTFITAV) CFR= 1.8*(DPFR-1.735E-04*MULTF(TIN)**0.2*W**.8/ROLIN4*DHLBR (RTL/LB)
C CIN= -TWCPI*G*(Ze+2.3)*DHLBR*CI/(TAU*3600.0032.17*144.0 (BTU/LB)
C CALCULATE CRAVITY AND FRICTIONAL PRESSURE DROPS IN NON BOILING, BI - 288 * QP * DISTR(NTS) / W (STU/LBO
C HEATED REGICN . . . . . . . . . . . . . . . . . . . . . . . . . 82 = -DhL8*(DPFR+DPG4I (BTU/LB)
C CDZDW - (-COHOW + CFR+ CIN)*ZRB / (81 + 82) (IFT)
DPGR = -RCLAV
* 811
/ 144.0 (PSI) C
PDZDW = CABSICCLDwI
DTFILM = (282.0*1.2916671*5P*SMULB**0.4/(G**O.8*KLF(TAV)**0.6
UPFR- (-3.3256E-11 *ZB8/O*FF(G*D/MUL8)*(,*G* (MULTF(TAV+DTFILMI/ ADZDW = -ATAN2(AIMAG(CDZOW),REAL(CDZDWIl/TWOPI
1 PULB)**0.14/ROLAV IF(AUZ'DW .LT. 0.0) ADD = 1.0 + ADZDW
PBBCAL * Pl + DPER + CPGR SDLON - TAU * ADZDW
DHLBB DhLFAPS8CAL) C
C C CALCULATE TFE INLET FLOW Ti ENTHALPY AT THE ACTUAL POSITION OF
W = C.7e54 * 0 * C * G C THE 8. 8. TRANSFER FUNCTION
. . . . . . . . . . . . . . . . . . .
CCMEGA = 3600.0*TWOPI*CI/TAU IE/HRI C
SPEHEA = IHLTF(TAV + 5.01-HLTF(TAV - 5.0)) * 0.1 STL/LB-- COHDWP = CDFIW + 81 * CDZDW / ZBB (8TU/L8)
HFC = 0.C279*W**O.8*KLF(TAV)**0.6*SPEHEA**0.4 / PDHDWP - CABS(CDHDWPl
E (0**1.d*MULB**O.4) ADHDWP = -ATAN2(AIMAG(CDHOWPI,REAL(CDHDWP))/TWOPI
C IF(ADHEDWP .LT. O.C) AO14OWP = 1.0 + ADHOWP
ABC - GP
/ h SDHDWP = TAU * ADHDWP
CI - CSQRT(CUMEGA/ALPHAl C
C2 = Cl * THICK * CI C CALCULATE TE PRESSURE DROP T.FS. IN THE SINGLE-PHASE REGION.
C3 = CI * CCOS(C2)/ (KW * CSIN(C2 * CE) C
C4 * Pt * D /(1t.0/HFC+C31wW*SPEHEA) CDPIDI - (-DPFR - DP(.R)*CUZDW / Z88 (PSI)
C5 - RCLAV*CCMEGA/G COPEFR = -CFR/DHLBB (PSI)
CK - CS + C4 CJP1IN - -CIN/DHLeS (PSI)
C6 - A * C3 /(1.0/HFC + C3) CCPL0G = COPIFR + CDPEIN (PSI)
CT = (1,0-Al * ABC CDPE - CCFICZ + CCPEIC (PSI)
CB = C& 8 ABC C
CL - C7 + C8 POPI - CABS(CUP0LI (PSI)
PDP1FR = CABS(CDPlFRl (PSI)
PDPIIN = CABS(CJPIIN) (PSI)
FIND IN WHICH TEST SECTION THE BOILING BOUNDARY IS LOCATED POP10G - CABS(CDPI0G) (PSI)
GENERATE FEATtD
AND UNHEATED SECTION LIMIT CDORDINATES AND CORRECTED POPE - CARSICDP I (PSI)
FEAT FLUX DISTRIBUTION AOP10Z - AMOU(I(.0-ATAN2(AIMAG(CDPIOZIRAL(CDP10ZI)/TWOPI),1.0 (CYCLES)
ACPlFR - AMCD((1.0-ATAN2(AIMAG(CDPlFR),REAL(CDPIFRI)/TWOPI .1.0 (CYCLES)
RATIO (CZU + DZH)I/CLH ADPAIN - ANCOiIE.0-ATAN2IAIMAG(CDP1IN),REAL(CDP1IN))/TWUPI),1.0) (CYCLES)
ZlE) = 0.0 ADPIDG - AMCD((1.0-ATAN2(AIMAG(COPODG),REAL(CDP10GI)/TWPI)e1.0) (CYCLES)
Z(21 = 0.5*DZU ADP1 = AMCO((1.0-ATAN2(AIMAG(CDP1 ),REAL(CDPI ))/TWOPI),1.0) (CYCLES)
D 88 1 2,14,2 C
DISTR(I) = QDISTR(I/2) * RATIO RE TURN
Zf1+1) - Z(I) + DZH END
IF(ZB-L(I+13) 881,8819882
SUBRCUTINE STAPR2(IMAX) Po(1) v PB8
C VERSION 2.4 10/16/69 ZO1) - ZoB
C X1i) * 0.0
C SUBROUTINE STAPR2 CALCULATES THE STEADY STATE CCNOITIONS IN THE P * ALCC(PBRI
C BOILING CPANNEL STARlING AT THE BOILING BOUNDARY AND MARCHING V() * 1.0/P0LFIP)
C DOWNSTREAM. Jul * G * V(1
C 'NLY FOR UNII-ORM HEAT FLUX DISTRIBUTION. oMI T
TP * t.0/ROGF(P) - V())I/HFGF(PBSI
C SEGMENTS OF EQUAL TRANSIT TIME ARE CONSIDERED. HE * HLF(PAB)
C XTTE * 1.OE+30
C REQUIRED INPUT I ALFAF 1.0E-10
C QP : LINEAR HEAT RATE (BTU/HR-FT) C
C G : MASS ILUX (LBM/HR-FT?) nO 999 1 - 1,59
C LRB POSITION 0F THE ROILING BOUNDARY (FT) HI * HE
C ZMAX : MAX VALUE CF COORDINATE ALONG THE CHANNEL TO BE PI * Pol)
C REACHED (FT1 PEFIN * Pt
C P88 : PRESSURE AT THE BOILING BOUNDARY (PSIA) 111 PE - PEFIN
C PEX : LIMITING EXIT PRESSURt VALUE (PSIA) El - (EXP(OMI*DT) - 1.0)/041
C UT : TRANSIT TIME INCREMENT (HR) DL - Jill * El
C HE a HI + QTP*DI/G
C OUTPUT : XE - (HE - HLF(PE)l/HFGF(PE)
C IMAX : AUMBER CF SPACE POINTS IF(XE .CE. 1.0) GO TO 997
C CM, 9,E : COEFFICIENTS DEFINED IN TEXT C
C X : JUALITY
VALUES C CALCULATE PRESSURE DROP USING THE LOCKHART-MARTINELLI CORRELATION
C 20: POSI'TICN ALCNG THE CHANNEL (FT) C
C J: VELOCITIES OF THE CENTER OF VOLUME (FT/HR) P * ALOGtPE)
C PC: PRESSURE VALUES (PSIA) ROL = RCLF(P)
C DPDLR : FRICTIONAL PRESSURE URUP SLOPE DIVIDED BY G**1.8 RUG - RlGrtP)
C (PSI/FT) / (LKM/HR-FT2)**1.8 MUL - MULFIP)
C OPDLGR : GRAVIATATIONAL PRESSURE DROP SLOPE (PSI/FT) MUG = MUGF(P)
C OPAC : ACCELERATION PRESSURE DROP VALUES CIVIDED G**2 RY C
C (PSI/(LBM/HR-FT2)**2) XTTI - XTTE
C V : FLOWING SPECIFIC VOLUMES (FT3/LBM) XITTE - XTTFIXEI
C DPULF - FI2LF)O.5*XTTI+XTTe)l*DPDL-F(O.5*(X(I)+XF)I
C X, ZO, J ANE PC ARE DEFINED EOR I = 1, IMAX UPOLFRII) - DPDLF/G**1.8
C 014, B, E AND THE PRESSURE DROP TERMS ARE DEFINED FOR I * 1, IMAX-1 ALFAI - ALFAE
C ALFAE = ALFAFIXTTE)
C PROPERTY FUNCTI-NS REUIRtO: ROLF, ROGF, MULF, MUGF, HFGF, HLF DPDLGR'(I) - DPDLGF(0.5*(ALFAI+ALFAE))
C DPACI) = 8GFIXE , ALFAE) - BGFIX(11, ALFAI)
REAL PUL,
MUG, MULE, r4UJF, J DP = IDPULF + OPDLGRII))*DL + DPAC(I)*G*G
COMMON// D, QP, G, Z8, ZMAX, PIN, P0B, PEX, TIN, DTSUBR, DT PEFIN * PI + UP
CCMMON /STA20T/ 0M(60),
B)60), C(60), X(60), ZO(60), PO(60), IFIPEFIN .LE.PEX) PEFIN = PEX
1 UPDLFR1601, UPOLGR(60), OPAC160), V(60), J(60) P = ALOG(PEFIN)
C VL * 1.0/FOLFIP)
C ********* FREON - 115 PROPERTY FUNCTIONS ****** VG 1.C/ROGF(P)
C V(I+1) - VL + XE*(VG-VL)
C THE ARGUMEAT CF RCLF, ROGF, MULF AND MUGF MUST BE ALOG(P) CMI = ALOG(V(1+1)/V)I))/DT
C IFIABS(PEFIN-PE/PEFIN - 0.002) 112,112,111
RCLF(X) =
.1033CPE 03 - .2657116E 01*X ROLF 112 PC(1+1) - PEFIN
I -
.3389485E -01*X*X - .830T981E -01*X*X*X ROLF OMI) * C1I
ROGF(X) EXP( - .326T308E C1 + .9181820E 00 *X ROGF ElI) *EXP(0MI*DT) - 1.0)/UMI
I + .3519715E-C2 *X*X ) ROGF Bl) = (EII) - DT)/CMI
MULF(X) * EXP( .1064657 E 01 - .3255410 E 00*X MULF X(1+1) - XE
1 + .6C64336 E-02*X*X - .1726090 E-02*N*X*X I ULF J(I1+1) = G*VfI+1)
MUGFIX) - .1967795 E-01 + .7899068 E-02*X MUGF ZO(I+1) - ZC(I) + DZ
1 - .5630545 E-02*X*X + .2141095 E-02*X*X*X MUGF IF(ZO(I+1) - ZMAX) 9S9,998,998
2 - .4536185 E-C*X*X*X*X + .1974262 E-04*X*X*X*X*X MUGF 998 IMAX
* I + 1
C RE TURN
HFGF(P) =* .7058670 E 02 - .9218673 E 00*P HFGF 999 CONTINUE
1 + .4283483 E-01*P*P - .1186139 E-02*P*P*P HFGF C IMAX - 1000 IS USED AS A CJDE TO INDICATE THAT EXIT WAS NOT REACHED
2 + .1250690 E-C4*P*P*P*P HFGF IMAX - 1000
C DTSKC = 360C.0 * DT
C HLF MUST BE PROVIDED EXTERNALLY WRITE(6,1) 2C(60), UTSEC
C I FORMAT(40************* STAPR2 ***** 60 ITERATIONS COMPLETED WITHOU
C PRESSURE CROP FUNCTICNS IT
REACHING ZMAX, Z0(60) -*, F10.5 /
C 2 * OT(SEC) * F10.5 //)
RELPF(XI G*(1.0-X)*QIMUL RETURN
DPDLFF(X = (-3.3256E-11)*FF(RELPF(X))*G*G*(1.0-X)*(1.C-X)/(RCL*D) 997 IMAX - I
DPDLGF(ALFA) =-U(ROG-ROL)*ALFA
+ ROL)/144.0 WRITE(6,2) IMAX, ZO(IMAX)
BGF)XALFA) = (-1.6628E-11)*((1.0-X)*(1.0-X)/((1.0-ALFAi*ROL) + 2 FORMAT(*0**********STAPR2 ******** A JUALITY LARGER THEN 1.0
1 X*X/(ALFA*RCG)) I , -WAS OBTAINED BEFURE REACHING ZMAX, 10(', 12, *1 -*,
XTTF(X) - (((1.0 -X)/X)**0.9)*((ROG/ROLI**0.5)*(MUL/PUG)**0.1 2 F8.5 //l
C RETURN
A - 0.7854 *0*0 FT21 END
QTP - QP/A B/HR-FT3
SUBROUTINE DYNPRL (KMAX)
C VFRSION 1.3 11/07/69 SUBACUTINE DYNPR2 (IMAX, KMAX, KMAX21
C SUbROUTINE UYNPRI CALCULATES THE PRESSURE DROP IN THE SINGLE-PHASE C VERSION 1.5 11/07/69
C REGION IA A NCK-LINEAR FASHION C SUBROUTINE DYNPR2 CALCULATES TIMt UEPENDENT PRESSURE DROP IN A
C C TWO-PHASE CIANNtL ACCJRDING TO THE ENTHALPY TRAJECTCRY MODEL
C Z11,K) ANF C(1,K) (-UR K =1,KMAX
ARE INPUT TO THIS SUBROUTINE C
C PI1,K) FOR K -1,KMAX IS THE OUTPUT C INPUT : Z(1,K), P(I,K), DiT(1,K), U(1,K),
K-1,KMAX : VALUES AT THE
C C STARTING POINT (USUALLY THE eg), AND
REAL MULB, MULTF, KLF, 4ULW C OM(I), E(I), B(11, PDLFR(I), DPDLGR(Il, DPAC( I1,
CCMMON // D, QP,GO, ZBB, ZMAX, PI , P88, PEX, TIN, DTSUBB, OT C FUR I = 1, IMAX - 1
CUM*CN /UYNPH/ Z(40,120), ZOT140,120), G(40,120), P140,120) C V(I1 FOR I =
1, IMAX AND OT. : REFERENCE STEADY-STATE
C C PARAMtTERS
C PROPERTY FUNCTICNS REQUIRED : ROLTF, MULTF,
KLF C
C REAL JI(12C), JLI(120, 021(120)
TIlT) = 1.0 / (T + 459.61 COMMON// D, QP,
GLZe8, ZMAX, P1 , PBS, PEX, TIN, CTSURB, OT
MULTF(T)= EXPI - .5483695 E 01 + .5585921 E 04*TIIT) MULTFI/2 COMMON /DYNPR/ Z(40,1201, aZDT(40,120l, G(40,120), P(4C,120)
1 - .1950534 E C7*TIIT)**2 + .3577018 E 09*TI(TI**31 MULTF2/2 CCMMON /STA20T/ CM(60), 6(601, E(60), X(601, ZO(60), PO(601,
ROLTF(T) - .1035453 E' 03 - .7105255 E-01*T - .6448694E-04*T*T ROLTFt/1 1 OPOLFR(60), DPDLGR(60), UPAC(60I, V(601, AJO(601
KLF(T) - 0.0475 - 6.76667E-05* T KLF 1/1 C
C 1MAXI a IMAX - 1
TAV - TIN + 0.5*DTSUBB KMAXI - KMAX
RCL - ROLTFITAV) DTINV - 1.0/0T
ROLIN - ROLTF(TIN) OTINCO - 1.0/((36CO.0*3600.0*32.17*144.0I*DTI
MULB - PULTF(TAVI C
DTINCC - 1.0/((3600.C*3600.0*32.171144.01*DT) C CALCULATE THE VOLUMtTRIC FLUX DENSITIES AT THE 8.B. . . . . . . .
DTFILM = (282.0*1.291667)*)P*MULB**0.4/IGO**0.8*KLF(TAV)**0.61 C
TWAV = TAV + DTFILM 00 89 K = 1,KMAX
MULW - MLLTFITWAVI 89 JI(K) G(1,IK) * VIl1
RE - GO*C/MULB C
DPOLGR = -RCL/144.0 C MAJCR LOOP
UPDLFR - I-3.3256E-11FFRE)*(G0**0.2*(MUL/MULSI**0.14/RCL*D C
C 00 999 1 *1,IMAX1
P802 = (DPDLFR*GOP*1. + CPDLGR)*ZBB + PI C
IF(A8S(Pd82-PBBf/PBB .GE. 0.02) WRITE(6,11 PB11,P882 C DETERMINE RELATIVE VEL3CITIES AT INLET
1 FCRMATIO0************STEADY STATE SINGLE PHASE PRESSURE DROP AS C
* ,'CALCULATED NY DYNPRI DOES NOT AGREE WITH SUBCBB VALUE :/ DO 99 K = 1,KMAXI
2 30X, 'P88(SUBCBR) =', F10.4, 1OX, 'PBBIDYNPRI) =0, F10.4, 99 JZIRI * Jt(K) - DZCT(I,K)
3 * PSIA' //) C
KMAXI - KMAX -1 C CAtCULATE-RASS FLUX, VELOCITIES AND COURDINATES OF MESH POINTS
DPI-ROI - -T.OIE-1O*MULTF(TIN)**0.2/ROLIN C
PIP = P1 - DPFRO1*GO**1.R KMAXI = KPAXI - 1
C C
DO 99 K = 1,KMAX1 DO 98 K = 1,KMAXI
G18 = GI1,K)**1.8 SIK - (JZI(K+1I - JZI(KI) * OTINV
DGOT - (G(I1,K+1-GTI,Kt*TINC1D DZI(K) aE It) * JZItK) + B(I) * SIK
OPTIFI - DPFR01*G18 - DGDT*2.3 C
DP - (DPDLFR*G18 + OPDLGR - OGDTI*Z(I,KI + DPO1FI C CALCULATE VALUES FOR NEXT SPACE STEP . . . . . . . . . , . . . . .
99 PI1,K) = PIP + UP C
RETURN JIfK) - JI(K+1I + CMIII * DZIfK)
kND C JIK) HERE IS REOEFINED AS J(I+.1,K)
G(I+1,K) - JIIKI / V(I+1)
Z(I+1,Kl = Z(I,K+1l + DZI(K)
98 DZDT(I+1,K) = DZOT(IK+1L) + E(I)*SIK
C
C CALCULATE PRESSURE DRCP AND PRESSURE . . . . . . . . . . . . . . .
C
KMAX2 = KPAXI - 1
GAVIKI = (G(1+1,1) + Gil,2))*O.5
C
00 97 K =1,KMAX2
GAVIK = GAVIKI
GAVIKI - (G(I+1,K+1) + G(I,K+2)I*0.5
DPIx = (OPOLFR(I)*(AVIK**1.8 + DPDLGR(II) - i6AVIK1-GAVIK)*DTIACC)
1 *DZIK) + DPAC(I)*GAVIK**2
97 F(I+1,IK - P(I,K+IJ + DPIK
999 CONTMNUE
RETURN -
END
SLBROLTINE PRTPL3(N3,t!, NM. NL,NS, KX,JX, YSC) VERSICN4 A(L)=A(LL)
12 AILL)=F
C PRTPL3 IS A MOCIFIEC VERSION OF
A SUBROUTINE SUPPLIED BY THE 14 CONTINUE
C INFORMATICN PROCESSING CENTER AT MIT. PRINTER PLOTS ARE PRODUCED 15 CCNTINUE
C AND THE EXTREMA OF TFE GIVEN FUNCTIUNS ARE DETERMINEO. 16 IF(NLL) 2C, 18, 20
18 NLL-50
DIPEASICA CUT(10I),YPR(I11I, IANG(9),A( 480), B(KX,JXI 20 WRITE 6,7)
INTEGER CUT, IANG/.,G'3,'4','5','6','7','8','9'/ BLANK=0
CCMlON/PLCTCT/ XMAXJ(9), YMAXJ(9), XMINJ(9), YMINJ(9), YRANJ(9) VER. 4.1 XSCAL=(AIN)-A(1))/(FLCAT(NLL-1)I
C YSCAL = C.01 * YRANGE
1 FORMAT(II, 60X,7H CHART ,13,/) IF(YSCAL .EQ. E.0) YSCAL = 1.0
2 FCRMATIIH, F11.4 , 5X, 101AX, 112) IF((YSC .GT. YRANGE) .ANU. (YRANGE .GT. YSC/10.0)) YSCAL=0.01*YSC
3 FORMAT(IH, 117X, 112) X8=A(1)
4 FORMAT('0PRTPL3 MIN ORDINATE =', F12.5, * AT ABSCISSA L=1
1 F12.5, * (CURVE NO-, 12, * ', IX, Al / DO 80 11,NLL
2 11X, 'MAX ORDINATE F',
12.5, I AT ABSCISSA F12.5, =, F=I-1
3 30X, 'ORDINATE RANCE =', F12.5) XPRTXO+F*XSCAL
7 FORMAT( 1H 16X,101H. IFIAIL)-XPR-XSCAL*0.5) 50,50,70
1 . . . . . .)
50 DO 55 1X=1,101
8 FORMAT(H ,9X,11F10.4) 55 CUT(IX)BeLANK
57 CONTINUE
IADV = 0 00 60 J=1,MY
IF(Y$C .GT. 0.0) IADV 1 56 LL=L+J*N
WRITE(6,1) IADV, NO JP=l(ALL)-YMIN)/YSCAL)+I1.0
MY = M - 1 OUT(JP)=IANC(J)
C 60 CONTINUE
C DETERMINE EXTREMA OFEACH CURVE. . . . . . . . . . . . . . . . . . IFIL.EQ.N) GO TO 61
C L=L+1
DC 45 J - 2,M IF(A)L)-XPR-XSCAL*0.5) 57,57,61
YMIN=1.CE75 61 CONTINUE
YMAX*-1.OE75 WRITE(6,2)XPR,IOUT( IZ), 1Z=1,101), I
DO 40 K = 1,N GO TO 8C
IF(8(KJ) - YMAX) 41,41,42 70 WRITE16,3) I
42 YMAX = B(KJ) 80 CONTINUE
XMAX = B(X,1) WR ITE(6,7)
41 IF(B(K,J) - YMIN) 43,40,40 YPR(1)YMIN
43 YMIN - B)I,J) DO 90 KN=1,10
XMIN = B(K,1) 90 YPR(KN+1)=YPR(KN)+YSCAL*10.0
40 CONTINUE WRITE46,8)(YPR(IP),IP=1,11)
XMINJ(J-1) = XMIN
XMAXJIJ-1) = XMAX WRITE THE VALUES OF TIHE MINIMUM AND MAXIMUM OF EACH CURVE
C
YMINJ(J-1) = YMIN C
YMAXJ(J-1) = YMAX 91 DO 92 J = lPY
45 YRANJ(J-1) = YMAX - YMIN 92 WRITE(6,4) YMINJ(J),XMINJ(J),J,IANu(J),YMAXJ(J),XMAXJJ),YRANJIJ)
RETURN
C PLOT THE CATA . . . . . . . . . . . . . . . . . . . . . . . . . *. END
IF(YSC .LE. 0.0) GO TC 91
YMAX=-1.0E75
YIN=1 .CE75
DO 49 J = 1,MY
IF(YMAXJ(J) .GT. YMAX) YMAX = YMAXJJ)
IF(YMINJIJ) .LT. YMIN) YMIN = YMINJ(J)
49 CONTINUE
YRANGE = YMAX - YMIN
1=1
DU 39 J=1,M
DO 39 K=1,N
A(I)=(K,J)
39 1=1+1
ALL=NL
IF(NS) 16, 16, 10
10 DC 15 I=1,N
DO 14 J=1,N
IF(A(I)-A(J)l 14, 14, 11
11 L=I-N
LL=J-A
DO 12 K=1,M
L=L+N
LL=LL+N
F=A(L)
.00 011111'
PRESSURE
RELIEF
VALVE
~R'
SER
.500
0-RING (EXAGGERATED)
BULK FLOW
THERMOCOUPLE
0.062" DIA,
WIRE BRAID POWER
LEAD
2kL~ it.%PVS
1 06
TRANSDUCER BRIDGE
ADAPTER
DIRECT
PREAMPLIFIER RITING
PSCILLOGRAPH
CALIBRATION
CIRCUIT
l
l 5k+l
REGULATED
POWER SUPPLY
/. Homogeneous Model
>
LM Freon -113 at 20 psia
L ock hart - Mart inellIi
G- 3.5 x I10 Ibm /hr ft"
G-IOxlO lbm/hrft"
D- 0.43 in.
0.01- l
0.0001 0001 001 0.1 1.0
QUALITY, x
FREON-i 13
D = 0.43 in.
z
LAL
0
z
0
0
C-)
Zn STAUB
= 2350
LEVY'S MODEL
2
q" = 2350 Btu/hrft
-20
2 5 10
[psi]
/ 18 1-7
/
6
APlex, [psia]
8
[psi] ATsubb 5
TOTAL PRESSURE DROP, APlex
7 50 r
66 40 .
5 30 / SUBCOOLING, AT subb
16 1-2
4 20 L /
pex = 14.7 psia
T1= 90 *F EXIT QUALITY, xex
3
q'= 400 W/TL
0
Uniform heat flux distribution
PRESSURE AT THE BB, pbb
0 No subcooled boiling
1000 1500 2000
MASS FLOW RATE, w [lbm/hr]
06
UNIFORM (1150
0 UNIFOR (96.0 *F
COSINE (94.8 *F)
c:)
Lu
LU
c-
0.0 . e
o 0
100
w 0
-- -
o 00 -
ISOTHERMAL (95.20 F)
In
a- 12 -
a. wA To r=94.6* F
0 Pc = 30.24 in.Hg
a o
a. 0- T~9. 0
- 00T
8 -3
=95.2* F
2.79
C. -0.008
44-
O. 2.18
0 -0.013v
BB
00 NVG
I-3
1. 64 -...
0.044 B
BB RUN NO 8
NVG oTo = 94.8 *F
1.250 0
0.084 0 NVGRB
B
1 2 3 4 5 6 7 8 ex
STATION ALONG THE CHANNEL
FIG. 4.4 PRESSURE PROFILES AT 4700 Btu/hrft 2, COSINE HEAT FLUX DISTRIBUTION
2 (o)
5.73
- 0.010
.-. , 4.48 13
0.011
(0)
3.70 V
0.030
0O
z
0
Ies
> 1.74 o 0 0
L. 0.157
it
RUN NO 6
W2 T. 94.6 *F
0.95
I 2 3 4 5 6 7 8BOx
STATION ALONG THE CHANNEL
FIG. 4.5 PRESSURE PROFILES AT 9400 Btu/hr ft, UNiFORM
HEAT FLUX DISTRIBUTION
RUN NO. 15 EFFECT OF NVG CORRELATION
T*' 115.2*F
8k
0.
N
0
2r 6 1-
0.
O I-- Staub
IIII I11II1 11 1
4 5 7 8ex
STATION NO. ALONG THE CHANNEL
2
FIG.4.6 PRESSURE PROFILES AT 14,100 Btu/hrft, UNIFORM
HEAT FLUX DISTRIBUTION
I I I I I I
C:
o A
o A
1.0 U AA
uj
O O v vT A
oy VA
0 o o v 6 v.
a- RUNS D4 TO D21 4-)~
L
AVERAGE POWER 0f 0
o 100 W/TL oL 0
A 300
OPEN POINTS : UNIFORM HEAT FLUX DISTRIBUTION O 400
CLOSED POINTS : COSINE HEAT FLUX DISTRIBUTION o 500 I VERAGE VALUES AND
STANDARD DEVIATIONS
0.1 0.2 0.3 0.4 0.5 0.6 0.7
EXIT QUALITY, xe
FIG. 4.7 ACCURACY OF THE PRESSURE DROP PREDICTIONS FOR ALL THRESHOLD AND TRANSITION POINTS
11
il1,,,,,llolmini lull
I I
6
o FIRST CYCLE
A SECOND CYCLE
O THIRD CYCLE
o FOURTH CYCLE
4 I MAXIMUM FLOW REACHED
CLOSED POINTS INDICATE
DECREASING FLOW
PRESSURE AT STATION I
2
N
w
co PRESSURE AT STATION 6
w O)
0r
0.4-
~ I I I 0
2 4 6 8 10
110 - Tcal*
U-
0
LLJ
LU
90
1 2 3 4 5 6 7 8ex
LU
S 150
LU
Tsat
130-
90 1 I I I al 1 1
1 2 3 4 5 6 7 8ex
STATION NO. ALONG THE CHANNEL
Tsat ; Saturation temperature based on measured pressure
T Calculated single-phase temperature
cal
T Measured temperature (Low temperature reading at
Station 8 is due to a thermocouple error)
FIG. 4.9 TEMPERATURE PROFILES FOR POINTS A AND B OF FIG. 4.8
WBMV Mwolll 11 M1911ill
1 .0__ _ ___ _
jQ*(z'jbj)
0.1
2 IH*(z,jw)I
0.001
0.0001 I I
0.1 1 10 100 1000
2ir
argQ*(z,jw)
3
S 21
di
di 1
2 2 argH*(z,jw)
Period, T [s]
Effect of T [s]
4-)
cnj
Effect of q'
rft]I
-[Btu/h
Reference point
zbb = 3.0 ft
Effect of G q = 800 Btu/hrft
G = 3.5x10 5 ibm
[105 lbm/hrft2] hr ft2
T 6 s
5/
I.e
Effect of z0
[ft]
[Btu/1 bm]
0.06
* 0.04-
0.02
0.00-
0 0 I | I I|
r~3
22
.-- 3,
C\j39
BB
0 2 4 6
Length along the channel, z [ft]
0.005
0.000 i i
2 '0,
2
N 344
s.-
C\j
BB
0 1 2 3
0.005
0.004
0.003
0.002
0.001
0.000
is M M
2w
1 2 3 4 5
0.08
0.04
0.00
3
2
. 0 1 2 3 4 5 6
0.004
0.002
0.000
0 1 2 3 4 5 6
Length along the channel, z Eft]
Experimental Point D18A-196; T- = 3.90 s; V0 = 0.291 ft/s;
A = 1.135 ft.
Cosine heat flux distribution shown on top.
0.002
0.000
27r
Tr
2
Tr
2
0 2 4 6 8 10
Length along the channel, z [ft]
The cosine distribution is simulated by N = 7, 14, and 28
segments. Two corresponding heat flux distributions are shown
at the top.
FIG. 5.11 EFFECT OF THE CHANNEL SEGMENTATION ON THE FLOW-TO-LOCAL
ENTHALPY TRANSFER FUNCTION.
- , , jP6
0 0.25 0.5 0.75 1.0 0 0.25 0.5 075
DIMENSIONLESS LENGTH , z DIMENSIONLESS TIME, t*
6w Re
friction
6Szbb
< 40
D17
unstable _ _ -- D10A
+ EXPERIMENTAL POINT
20 D9 THRESHOLD POINT
I TRANSITION POINT
o INTERPOLATED THRESHOLD
D6 0,1, etc
ORDER OF OSCILLATIOt
D4 D21 RUN NUMBER
2TR2B
0 0.2 0.4 0.6 0.8 1.0
q/w h
0
FIG. 7.1 STABILITY MAP AT 100 W/TL, UNIFORM HEAT FLUX DISTRIBUTION
lill N~~ m milii , .'i,,WimuI l
40
H- 40
20
0
0.2 0.4 0.6 0.8 1.0
q/w0 hfg
FIG. 7.3 STABILITY MAP AT 300 W/TL, UNIFORM HEAT FLUX DISTRIBUTION
1 I1li 1l haia
u u'lll
liilll
lll1ll lllin llull
i lll ..i
lll0it
ll
IIMI.l.. ill
liiiiii ll4
l i t Ill ' ill blii i i il A i il II I
STABLE REGION
D23
40 I- D22 +0
\ D11
20 F
FIG. 7.5 STABILITY MAP AT 200 W/TL AVE., COSINE HEAT FLUX DISTRIBUTION
W**mp"wWW 9.001*0.0
Run TR8, 300 W/TL, unif. &Tsubb=6 5 .40F
Fundamental
we 3 mode
30s
Third-order
oscillation
fwadt
wI ___Is
~' _
B Run Di0A
200 W/TL
[lbm/hrj
00
0
-4-
AO unif.
Points 93
and 93A
Run D16
Idiv = I 200 W/TL
unif.
Points 20
and 21
o/
O 60 G/
0/ A ~ (200)
c13
0
u>
0
co~40 (200)
U)
/ POWER LEVEL
/ A 100 W/TL
0 200
/ O 300
V 400
A 0 500
0
A 380 - -
q / Wohfg
Open points: First occurrence of oscillation, zero order.
Closed points: First occurrence of oscillation , higher-mode
(Number of tails denotes order of oscillation).
Points with a bar: Transitions to fundamental model.
Crowley, Deane and Gouse's boundaries also shown.
100
m500
r__-
\
0
\. STABLE REGION
\ \.
N
380
C)
C)
C..)
I I
-LJ
* I. I
* I
I I
I I
*1 )
~-
I I I a a
300
80 I \
'
20
660
200 STABLE
REGION
I
-J
C-)
100
//
'I
20 I I /
/
% \I' /
POWER DISTRIBUTION
COSINE
I'
/ I ------ UNIFORM
------ I
06
A0
A0
9 0200
0.4 -A AA UNSTABLE
oil REGION POWER LEVEL
A (-- W/TL
c3 200
0 300
v 400
380 C0 500
(c-D-G) 680 CDG38050
0.2
(C-D-G) 500 Closed points indicate transitions
(c-D-G) 600 ( higher zero-order curve at 200 l
see Fig. 7.2)
II I I
0 0.2 0.4 0.6 0.8 1.0
q/w0 - hfg
8 -
0v
Lii
4 - A POWER LEVEL
A 100 W/TL
m 200
* 300
Open points denote V 400
transitions * 500
0 1 1
0 20 40 60 80
SUBCOOLING, ATsubb [*F]
TO00
---- UNIFORM HEAT FLUX
DISTRIBUTION
81-
.- -V 400
POWER LEVEL
41-
* 200 W/TL
* 300
V 400
ATsubb
[OF] 0 F]
T = 3
40-
20 2.5
W uniform
0.5
exact cosine
27r
uniform resulting
difference
3 in delay
3
exact cosine
SUBCOOL ING
FIG. 7.15 ENTHALPY AND SINGLE-PHASE DELAYS - UNIFORM VERSUS COSINE HEAT FLUX DISTRIBUTION
Run TR6 Run TRiA
300 W/TL,unif. 300 W/TL,unif.
ATsubb=3 4 .30 F Is 9
ATsubb= .3-5.40F
111we
I s - -0
F Run D21
Run E02 psi Psnt 2 0
30W/TL,COS
Point 6
Tsubb= 76. 2*F
AT
Wo
[ lbm/hr]
80 - ---
0- 6
60 E03 (all') -----
~~
60 E-0TL
(al)0
40 E06 (-3 TL
1.0
-0 CORRECTED ORDER OF OSCILLATION
........... STABLE
THIRD-ORDER
0.8
-0 SECOND-ORDER
FIRST ORDER
200 W/TL
-0
-o
-0
-o
U
0.6 REFERENCE POINT
Ix
0.4
E05 (-2 TL)
E03 (all)
6Ap = 0
1.2 A
A
1.0 | - -A . A D
Wc3
0.8
ci pA 20W/T7
A200 W/TL
0.6
UN4IFORM HEAT FLUX DISTRIBUTION 03 300 W/TL
0.4
8s
W T ZOb/V
w0 1 bb o
Curve Point [lbm/hr] [s] [s]
A 012-01 Stable 547 - 0.960+ Plotted values
B 012-92 Threshold 410 2.28 0.805 multiplied by 2
C 012-03 Unstable 310 2.44 0.710
D D12-06 Unstable 2'6 2.43 0.426
Subcooling, ATsubb = 8.5 to 5.7 F ; Power, q = 300 W/TL, unif.
I I'Re
B
35
Io 5
8
'.5
.7.
1.0,
45
5.5
wW0 t zt~
Zb/Voo
Curve Point [lbm/hr] [s] [s]
A D16-12 Stable 445 - 4.80
B 016-93 Threshold 225 7.75 5.07
C D16-14 Unstable 197 7.75 5.18
Subcooling, ATsubb = 51.5 *F ; Power, qw = 300 W/TL, unif.
The curves are graduated in periods [s].
FIG. 8.5 OPEN LOOP TRANSFER FUNCTIONS AT HIGH SUBCOOLING