MITNE-1 92

MATHEMATICAL MODELS FOR

PREDICTING THE THERMAL PERFORMANCE
OF CLOSED-CYCLE WASTE HEAT
DISSIPATION SYSTEMS
by

ERIC C. GUYER
and
MICHAEL W. GOLAY

October, 1976

DEPARTMENT OF NUCLEAR ENGINEERING

MASSACHUSETTS INSTITUTE OF TECHNOLOGY
77 Massachusetts Avenue
Cambridge, Massachusetts 02139
Massachusetts Institute of Technology
Department of Nuclear Engineering
Cambridge, Massachusetts

Mathematical Models for Predicting the Thermal

Performance of Closed-Cycle Waste Heat
Dissipation Systems

by

Eric C. Guyer and Michael W. Golay

:1

ABSTRACT

The literature concerning the mathematical modelling

of the thermal performance of closed-cycle waste heat dissi-

pation systems for the steam-electric plant is critically

examined. Models suitable for survey analysis of waste

heat systems are recommended. The specific models discussed

are those for a mechanical draft evaporative cooling tower,

a natural draft evaporative cooling tower, a spray canal,

a plug-flow cooling pond, and a mechanical or natural draft

dry cooling tower. FORTRAN computer programs of these models

are included to facilitate their application.

2

TABLE OF CONTENTS

Page
ABSTRACT I

TABLE OF CONTENTS 2

LIST OF FIGURES 3
LIST OF TABLES 4
CHAPTER 1. INTRODUCTION 5
CHAPTER 2. MATHEMATICAL MODELS FOR PREDICTING THE
THERMAL PERFORMANCE OF CLOSED-CYCLE WASTE
HEAT DISSIPATION SYSTEMS 6
2.1 Introduction 6

2.2 .Mechanical Draft Evaporative Cooling Towers 9

2.2.1 Literature Review 9
2.2.2 Selection of Model 13
2.2.3 Application of Model 18

2.3 Spray Systems 21

2.3.1 Literature Review 21

2.3.2 Selection of Model 23

2.4 Natural Draft Evaporative Cooling Towers 26

2.4.1 Literature Review 26

2.4.2 Selection of Model 28

2.5 Cooling Ponds 33
2.5.1 Literature Review 33
2.5.2 Selection of Model 35
2.6 Dry Cooling Towers 37
APPENDIX A. COMPUTER PROGRAMS FOR WASTE HEAT REJECTION
SYSTEM THERMAL PERFORMANCE CALCULATIONS 45
3

LIST OF FIGURES

2.1 Illustration of Tower-fill Finite-Difference Calcu-

lation 15
2.2 Calculational Algorithm for Prediction the Perfor-
mance of Mechanical-Draft Cross-Flow Evaporative
Cooling Tower 17
2.3 Comparison of Reported and Predicted Mechanical
Draft Cooling Tower Performance 20
2.4 Computational Algorithm for Spray Canal Thermal
Performance Model 24
2.5 Calculational Algorithm for Natural-Draft Evaporative
Cooling Tower Performance Model 29
2.6 Comparison of Reported and Predicted Natural-Draft
Cooling Tower Performance 32
2.7 Cooling Pond Model Computational Algorithm 38
2.8 Dry Cooling Tower Schematic Drawing 40
LIST OF TABLES

A.1 Required Data Input for Natural-Draft Evaporative

Cooling Tower Model 46

A.2 Required Data Input for Spray Canal Model 47

A.3 Required Data Input for Mechanical-Draft Evaporative

Cooling Tower Model 48
5

CHAPTER 1

INTRODUCTION

A study of mixed-mode waste heat dissipation systems

for steam-electric plants has been performed in the Nuclear

Engineering Department [28]. Mixed-mode waste heat dissipa-

tion systems are defined as those waste heat dissipation

systems composed of combination of two or more different

heat rejection devices or those waste heat dissipation sys-

tems operated with variable cycles. This report summarizes

the mathematical thermal-performance models of the various

component waste heat systems which were developed in the

completion of this study. The thermal-performance models

presented in this report are those for a mechanical-draft

evaporative cooling tower, a natural-draft evaporative

cooling tower, a spray canal, a plug-flow cooling pond, and

a natural or mechanical draft dry cooling tower. These

models are recdmmended for survey-type analyses of waste

heat rejection systems.

Chapter 2 includes a review of literature pertaining to

the mathematical modeling of each of these heat rejection

devices and a discussion of the assumptions inherent to the

recommended models. Appendix A contains FORTRAN program list-

ings of the recommended models and a description of the re-

quired input data for each of the models.

6

CHAPTER 2

MATHEMATICAL MODELS FOR PREDICTING THE THERMAL

PERFORMANCE OF CLOSED-CYCLE WASTE HEAT
DISSIPATION SYSTEMS

2.1 Introduction
The literature concerning the dissipation of waste heat
from central power stations has grown rapidly in the last

decade. All areas within the general category - from biologi-

cal effects to heat transfer developments - have been the

subject of an increasing number of technical reports, journal

articles, and trade magazine articles.

The two fundamental reasons for the rapid growth of
this literature are the imposition of.environmentally-motivat-
ed governmental regulations on the traditional "once-through"

cooling system and the increasing unavailability of adequate

sources of "once-through" cooling water at otherwise attrac-

tive central power station sites.
However, there is as yet no definitive source of informa-

tion from which one can independently construct reliable

thermal behavior and economic models of waste heat dissipa-
tion systems. The few studies which have addressed the gener-

al problem of developing the independent capability of evaluat-

ing the thermal performance of alternative waste heat.dissi-
pation systems are either out-of-date [9] or lacking in the
?

details 13] [10] and thus c'an not be directly applied to

the present task. Thus, considerable effort was required

to review the available information and compile it into a

useful tool for evaluating the costs/benefits of various
alternative waste heat dissipation schemes.
The available literature concerning the mathematical
modeling of the economics and thermal behavior of waste
heat systems has been authored primarily by 1) the vendors
of waste heat dissipation equipment, 2) the electric utility
incistry, and 3) various research institutes and universi-

ties. In view of the present task of developing accurate

mathematical models of conventional waste heat rejection

devices some general comments can be made about the litera-

ture with regard to its authorship.
Although there has been a tremendous increase in the

waste heat dissipation equipment vendor sector in both size

and diversity, the publications of these vendors are gener-

ally qualitative in nature. With a few notable exceptions,

the literature published does not deal quantitatively with

thermal behavior analysis, but, rather, describes qualita-

tively the particular vendors present capabilities and high-

lights the economic advantages of the particular vendors

devices. Little of this information is of value to those

interested in developing an independent analysis capability.

8

The dearth of substantial information published by equip-

ment vendors is, of- course, understandable since their

proprietary interests are not well served by the free-flow

of their costly research and development results..

The literature on this topic authored by the electric

utility industry has come from the electric utilities them-

selves as well as their consultants - mainly the large

architectural engineering firms. As is the case above,

little substantive information has been published with re-

gard to the mathematical modeling of the thermal behavior of

various heat rejection systems by this sector. However,

valuable government-sponsored information has been reported

by architectural engineering forms. Many trade journal

articles which review the waste heat dissipation solutions

applied to specific sites have been authored by utility sys-

tem engineers, but these findings are usually of little

value to the present task.

Much useful information concerning the mathematical

modeling of the thermal performance of heat rejection sys-

tems has been authored by various research institutes and

universities under the sponsorship of federal and state

agencies and electric utilities. In applying some of this

information, however, difficulty is encountered in attemp-

ting to relate the published results to the actual thermal

9

performance of modern, well-designed waste heat dissipation

systems.

2.2 Mechanical Draft Evaporative Cooling Towers

2,2.1 Literature Review

Croley et al. [2] have recently addressed the problem

of developing an accurate thermal and economic model of

conventional cross-flow mechanical draft evaporative cooling
towers. Their review of the literature led then to the use

of a thermal analysis model based on a simple straightfor-

ward finite-difference solution of the well-known Merkel [11]
evaporative 'heat transfer differential equation.
The Merkel formulation of evaporative heat transfer

combines the mass transfer (evaporation) and the sensible

heat transfer coefficient into a single coefficient. The

approximate net energy transfer is then a product of the

coefficient and the enthalpy potential difference between
the water and the air streams. The standard "Merkel" equa-

tion is as follows:
T1
KaV
L J T2
dT
(h"-h)
(2.1)

where K = overall transfer coefficient, lb/(hr)(ft2 of

interface)(lb of water/lb of dry air)

a = interfacial contact area (ft2 /ft 3 of tower fill)

10

V = planar volume It 3/ft 2 of plan area,)

L = water. flow rate(lb/hr-ft 2 of plan areal

T inlet water temperature, and

T2= exit water temperature.
dt water temperature differential
h"= enthalpy of saturated air at the water
temperature
h = enthalpy of the main air stream(BTU/lb of
dry air)

Derivation of this relationship may be found in several

references[26] [27]., Physically the quantity KaV/L in the

above equation represents an effective heat transfer ability
or "number of transfer units" for a particular cooling tower.
This coefficient is dependent on the relative amounts of
water and air flow in the tower and must be determined experi-

mentally.
Croley et al. [2] have applied this differential equa-

tion in finite-difference form to solve the two-dimensional

heat exchange problem of the widely-utilized induced draft
crossflow evaporative cooling tower for known inlet air and

water boundary conditions. The finite-difference approxima-

tion to the Merkel equation consists basically of the divi-
sion of the energy transfer volume into a number of equal
sized blocks over which the energy transfer potential
11

(enthalpy) is averaged.
The conclusions of Croley et al. concerning the utility.

of the .basic Merkel formulation for the predicting of the

energy transfer in a cooling tower has since been substanti-
ated by the recommendation of Hallet [12]. Hallet, repre-
senting a leading cooling tower vendor, has suggested that
the best approach (for a non-vendor) to the problem of
evaluating the thermal performance of wet cross-flow towers
is a finite-difference solution of the basic Merkel equation.
This author also points out that, although many improvements
in the theory of simultaneous heat and mass transfer at

water/air interfaces have been suggested, the basic Merkel

formulation is the only widely accepted and proven theory.

The analysis technique suggested by Hallet is essentially

identical to that of Croley et al. except that Hallet recom-

mends the inclusion of a temperature dependence in the expres-

sion for the tower fill energy transfer coefficient:

Ka = f(T1 ) (2.2)

where T is the tower inlet water temperature. It. is interes-

ting to note that no physical justification is given by
Hallet for this "temperature effect". Consideration of
recent works which address the errors inherent to the Merkel

equation suggest that this "temperature effect" fixup is

12

necessary because of errors- in the Merkel approximate formu-

lation for evaporative heat transfer.

The investigations of Nahavandi [13] and Yadigaroglu

[14] have been concerned with an evaluation of the errors in-

herent to Merkel equation. The results of Yadigaroglu are

based on a comparison of the predictions of the Merkel

theory and a more exact and complete theory which treats the

mass and sensible heat processes separately. This investi-

gator found that the effect of the various approximations of

the Merkel theory tends to be small since the different approxi-

mations of the Merkel theory result in partially cancelling

positive and- negative errors. The conclusion is that, given

the other errors associated with cooling tower performance

predictions (uniform air and water flow rates, for example)

and performance verifications (experimental uncertainties),

the added complexity of performing the more exact energy

transfer analysis is not justified. Nevertheless, it is of

interest to note that Yadigaroglu found that the net positive

error in predicting the cooling range increased with increas-

ing air inlet temperature and humidity. This error could be

corrected by arbitrarily decreasing the value of KaV/L by

the appropriate amount as the water inlet temperature increas-

ed. Indeed, this is the same, but unjustified, approach

recommended by Hallet. Examining the magnitude of the over-

i3

prediction resulting from the use of Merkel theory ( on the

order of 5%), it is found that the Merkel theory error is

consistent with the suggested "temperature effect" correc-

tion of KaV/L (about 5% per 10 OF rise in inlet water temp-

erature for inlet water temperatures in excess of 90 OF).

2.2.2 Selection of a Model

The mathematical model to be used in the prediction of

the thermal performance of mechanical draft evaporative

cooling towers is the finite-difference approximation of the

Merkel equation. The finite-difference approximation to the

Merkel equation can be stated as [2]

rI I
KaV h i-h + h -h0
G(h0-hi) = N - 2 (2-3)

where h and h = saturated air enthalpies at the inlet

and outlet of an incremental element,

h and h = saturated air enthalpies at the temp-

erature of the water entering and
leaving the incremental element,
G = air flow rate per incremental element,
N = square root of the number of incremen-

tal elements, and

KaV = transfer coefficient.
14

The application of this equation to the cross-flow prob-

lem of a conventional induced draft cross-flow cooling tower

is shown in Fig. 2.1.

In addition to the above equation, the energy balance

equation

G(h -h ) = cpL(t -t = L(h -h0) (2.4)

is needed to completely describe the temperature history of

the air and water as it passes through the tower fill. In
the above equation:

L = water loading per incremental element,

t and to inlet .and outlet water temperatures for an

incremental element, and

c = specific heat capacity of water.

Equations 2.3 and 2.4 form a set of coupled equations

with unknown variables h0 and h0 which must be solved for
iteratively. The algorithm for calculating the average out-.
let water temperature and average outlet air temperature is
given in Fig. 2.2. Note thta, for practical purposes, the
water and air flow rates are fixed by the tower design and
to a good approximation can be assumed to be uniform and

constant throughout the tower. Note, also, that the algorithm

is for calculating the performance of a given tower design.
If we wish to find the size of the tower needed to meet a
15
Fig. 2.1

Illustration of Tower Fill Finite-Difference Calculatton

unit

Tower Fill
Structure

lenght

L, hl

4'
Unit Cell
ho G,hI
H= N
A~Y
H=N
EX

1L
hK
1-6

specific cooling requirement, a trial and error calculation

may be performed.
The saturated air enthalpy used as the driving poten-
tial in the Merkel equation depends on both the dry bulb
temperature and the humidity of the air. However, a good
approximation to the enthalpy which depends solely. on the

thermodynamic wet bulb temperature may be derived. From

Marks [15) we have the relationship,

E = 0.24Td + w(1062.0 + O.4 4Td) (2.5)

and

W* - (0.24 + 0.44W*)(Td - Twb)

W = (2.6)
(1094 + 0.44Td - Twb)

where E = enthalpy of moist air,

Td = dry bulb temperature,

W = specific humidity,
Twb = wet bulb temperature, and

W* = specific humidity for saturation at Twb.

Substituting the latter into the former we have

E = 0.24Td + W*(1062 + .44Td) -

(0.24 +0.44W*)(Td-Twb)(l062+ 0.44Td) (2.7)

(1094 + 0.k44Td-Twb)(
17

Fig.- 2.2

Calculational Algorithm for Predicting the Performance

of Mechanical Draft Cross-flow Evaporative
. Cooling Tower (MECDRAFT Program)

Determine inlet air and water enthalpies

Iteratively calculate h' and h0 for each

element of the top row starting at the
air inlet side (equations 2.3 and 2.4 )

NO
-4
For the next lower row of elements iteratively
calculate ho and ho starting from the air in-
let side (equations 2.3 and 2.4) I
4
Determine if bottom of tower attained I
I.

YES

Average outlet air and water temperatures for

each horizontal and vertical row respectively

End
18

Now assuming that in the deminator we can make the approxi-.

mation

32 - Twb 0 (2.8)

and expressing the saturation himidity in terms of satura-

tion pressure we have

0.6 22P
E z 0.24Twb + Pa(1062.0 + 0.44Twb) (2.9)
Patm Psa

where Patm = total atmospheric pressure, and

Psa = saturation pressure of water vapor at Twb.

The above assumption is a good one in this particular

circumstance since the error.affects the ratio of large num-

bers. An error of 50 *F in magnitude in the denominator
would be typical with the total resultant error being about

5%. However, in all applications of the approximate enthalpy

equation the equation is ultimately used to find the differ-

ence of two enthalpies and thus the resultant error in the

difference is minimal.

2.2.3 Application of Model

To achieve the goal of obtaining an accurate thermal per-

formance model of a conventional cross-flow induced draft

evaporative cooling tower module the physical dimensions and

empirical heat transfer and air friction data for a typical
19

module must be acquired. Croley et al. [21 have modeled

the thermal behavior of such modules and reported the results.

From the published information the physical dimensions of

the tower fill are readily obtainable. They are

height = 60 feet

width.= 36 feet, and

length = 32.

However, the air friction factors for this fill is not direct-
ly obtainable from the published results. Nevertheless, an
energy balance on the modeled tower based on the published

information indicates an average air flow rate of 2.4xlO3 lbm/

hr-ft 2 . It will suffice for the purposes of this study to
assume the air flow is constant and equal to this value.

Croley et al. do not report the values of the energy transfer

coefficient used in their study since empirical proprietary
information was used in evaluating the energy transfer coeffi-
cient. However, sufficient calculational results using this
proprietary information are reported to allow a regression
of the required information.

The Cooling Tower Institute [16) states that the depen-

dency of the energy transfer coefficient Ka on the air and

water flow rates in a tower can be well expressed by a rela-

tionship of the form

20

Fig. 2.3
Comparison of Reported and Predicted Mechanical Draft
Cooling Tower Performance

Water Loading = 6200 ibm/hr - ft 2

Air Loading = 1692 lbm/hr - ft2

95

90|
Tower
Discharge
Temp.(OF)
85

X -reported
80
- predicted

75

p p

65 75
Wet Bulb Temperature (OF)
21

Ka = aGs L 1 . (2.10) -

where a depends on the fill configuration and 8 is, to a

good approximation, equal to 0.6. Using the following

expression

a = 0.065 - (T1 - 110.0)*(0.000335) T 1 >90*F

and (2.11)

a = 0.0715 T. 900 F

where T 1 is the inlet water temperature, the performance

predictions of Croley et al. based on proprietary data can
be closely matched as shown in Fig. 2.3. This value of a

is consistent with the type of fill used in modern towers

and the values of a experimentally determined by Lowe and
Christie [23].

2.3 Spray Systems

2.3.1 Literature Review

Spray cooling systems for the dissipation of waste heat

at large central power stations are a relatively new con-

cept 17J. As a consequence, the development of thermal
analysis techniques for these systems is presently incom-
plete. The development of reliable mathematical prediction

models has not been achieved and has been hindered by the
complexity of the problem.
22

As opposed to cooling towers, the water-air interfacial

area and relative air to water flow rates are not well
defined for spray systems. Open to the atmosphere, varia-
tions in the ambient wind result in different spray patterns,
different air flows through the sprays both in magnitude

and direction, and different interference effects between

the individual sprays. The spray canal system also presents
a channel hydraulics problem in that the behavior of the
water in the canal must be understood to insure optimum spray
system performance.
Porter et al.[18] [19] have authored the only two present-
ly available detailed works on the thermal performance of
spray canals. The two papers represent two different approach-
es to the problem, one analytical and one numerical. Both
models, however, are based on the same limited data which
according to the authors result in optimistic predictions [20]-
Richards of Rockford [4] have published some limited

information concerning the application of their spray modules.

They indicate that an empirical "NTU" approach is used in
the basic heat transfer calculation. Most interesting, how-
ever, is their description of the flow requirements of the
channel in which the spray modules are utilized since this

description indicates their recognition of the importance of

the channel thermal-hydraulics in the overall performance of
the system.
23

2.3.2 Selection of Model

For the purposes of survey-type analyses, the numerical
prediction of the thermal performance of spray canals as
suggested by Porter et al. [19] is most advantageous. In
this model the heat transfer ability of each spray module
is defined by an empirical "NTU" or number of transfer units

which is dependent on the ambient wind speed. The effects

of air interference between individual sprays is considered
through the use of an empirical air humidification coeffi-
cient. Given the ambient meterological conditions rad inlet
water temperature and flow rate, the calculational procedure
is to march down the canal taking into account the cooling

effect of each spray module as it is encountered. The basic

calculational algorithm is given in Fig. 2.4.

. The heat transfer equation used in the model is

C (T - T )
NTU= ( n (2.12)
(h(T,.) + h(Tn
2 - h(Twb)

where C = specific heat capacity of liquid water,

p
T = temperature of water exiting spray nozzle,
n
Ts = final spray temperature,
h(T) = total heat or sigma function as defined by

Marks [15]1,
Twb = local wet-bulb temperature, and
24

Fig. 2.4

Computational Algorithm for Spray Canal Thermal

Performance Model (SPRANAL Program)

Pass = 1

----- Calculate heat transfer for upwind module

4,
r4 Calculate heat transfer for next downwind module -A
No

I Determine if heat transfer calculation completed

for all sprays in pass I
yes
j
Mix cooled water with main stream to obtain
spray inlet condition for next pass

Subtract evaporated flow from total flow

No
Determine if end of canal attained

yes

end
25

NTU = number of transfer units of an individual

module.

The total heat or sigma function used as the driving

potential for the energy transfer is defined by Marks [15] as

E hm - W hf (2.13)

where h = enthalpy of moist air at the wet bulb

temperature,

W = specific humidity for saturation at the wet

bulb temperature, and

hf.= enthalpy of liquid water at the wet bulb

temperature.

However, comparison of the sigma function and the enthal-

py indicates that, for the temperature range and temperature

differences of interest the following is a good approximpation;

AE(twb) Z Ah(Twb) (2.14)

where h is the enthalpy of saturated air at temperature Twb.

Since we are attempting to determine Ts by using Eq.(2.12)
and Ts is a term in the same equation an iterative solution
is necessary. The evaporated water loss is calculated using
the expression of Porter [19] It is
26

a = Cp(T - T Vig(i + B) (2.15)-

where a = fraction of water evaporated in each spray,

g = specific heat of vaporization of water, and

B = so-called Bowen ratio of sensible to evapora-

tive heat transfer.

In the application of the above equations, the Bowen

ratio can be conservatively set equal to zero, since, in
any case, the effect of water evaporation on the spray canal
thermal performance is minimal.
From the data given by Porter the relationship between
the NTU and windspeed has been deduced to be approximated by

NTU = 0.16 + 0.053*V (2.16)

where V is the windspeed in miles per hour.

In this model no direct account is made of the thermal-

hydraulic behavior of the water in the channel. However,
Porter has made some simple arguments in favor of assuming
that the channel is vertically fully-mixed between successive
passes of sprays.

2.4 Natural Draft Evaporative Cooling Towers

2.4.1 Literature Review

Conceptually, the thermal analysis of natural draft evapor-

ative cooling towers is a straightforward extension of the
27

mechanical draft cooling tower analysis developed in this

chapter. However, from a practical standpoint the problem

is considerably more complex since the heat transfer char-

acteristics and the air flow in the tower are dynamically

coupled. Also, in addition to needing to know the empirical

heat transfer coefficient of the fill, one also needs to know

the empirical air friction factors for the tower structures

and the fill. Further, a more exact determination of the

psychrometric condition of the air exiting the fill is desir-

abXe since this condition ultimat'ely determines the overall

performance of the tower.

In the past, attempts have been made, notably by Chilton[21]

to simplify the performance prediction for natural draft

evaporative cooling towers by applying an empirical relation-

ship for the overall thermal behavior. These efforts, how-

ever, were-not well received and presently the suggested

approach to the thermal analysis problem is based on a detailed

evaluation of the important physical phenomena.

Keyes [22] has outlined the necessary steps for the con-

struction of a thermal behavior model of natural draft cooling

towers. Essentially, the mathematical modeling of a natural

draft tower requires the solution of three coupled equations.

The equations are 1) an energy balance between the air and

water streams, 2) an energy transfer equation for the combined

evaporative and sensible heat transfer, and 3) an energy

28

equation for -the density induced air flow through the tower.
Keyes only reviews the general problem and discusses the

empirical information which is available for accomplishing

the modeling task.
Winiarski et al.[24] have developed a computer model

of the thermal behavior of a natural draft cooling tower

based on the three equations mentioned above. The author

notes, however, that the model presented awaits final verifi-

cation based on reliable test data from actual towers.

2.4.2 Selection of Model

The model of Winiarski et al. [24] has been chosen as the
basis for the development of a thermal behavior model of
natural draft evaporative cooling towers. The thermal ana-

lysis calculational procedure is reported in the form of a

computer program. The basic computational algorithm is given
in Fig. 2.5. The major remaining task in the model development
was, thus, the acquisition of the necessary empirical informa-
tion which would enable the computer program application. In
this regard all domestic vendors of natural draft evaporative

cooling towers were contacted and sufficient information was

obtained.
The data obtained was not typical heat transfer coeffi-
cients and air flow friction factors for a modern natural
draft tower but instead consisted of a set of typical perfor-
2Q

Fig. 2.5
Calculational Algorithm for Natural Draft
Evaporative Cooling Tower Performance
Model (NATDRAFT Program)

Input data including:

atmospheric conditions
packing characteristics
desired tower height
desired inlet water temperature
water loading

Estimate air flow rate

Calculate friction coefficient

Calculate heat transfer coefficient I

Estimate outlet water temperature

6

Counterflow integration scheme

No
-,Is inlet water - desired value?
Yes f
Calculate pressure losses

No
Is calculated H - desired value? -I
Yes
Resulting output describes tower performance

END
30

mance curves and tower and fill structural dimensions. Thus,

is was required to fit the computer model to the performance
curves.by a trial and error selection of appropriate heat
transfer coefficients and friction factors. The performanoe

data are known to be based on roughened-surface parallel-

plate-type tower fill with counter air/water flow. Rish [25]
has reported an empirical relationship for the heat transfer
coefficient and friction factors for smooth parallel plate
packing. They are;

C = 0.0192(L/G) 0 -5 , (2.17)

and.

h C pC fG(21)
L=-0.25 (2.18)
2 + 71. 6 Cf($)

where Cf = friction factor,

Cp= specific heat capacity of liquid water,

G = air flow rate lbm/ft -hr,

.L = water flow rate lbm/ft 2 -hr, and

h = heat transfer coefficient for evaporative
and sensible heat transfer based on enthalpy
difference potential.

It was assumed that the effect of the roughened surface of

31

the parallel plates could be simply accounted for by a fric-

tion factor multiplier F . That is;

Cfa =F m Cf (2.19)

where Cfa is the actual friction factor. The relationship

between the heat transfer coefficient and the friction factor

was assumed to remain the same.
A trial and error approach to determining Fm was used

and, as Fig. 2.6 indicates, a value of F of 3.2 gives excel-

lent results over a represeAative range of operating tempera-

tures and flow rates. In the determination of Fm all other

air friction effects other than that of the fill were neglected.
All the details of the computer model will not be dis-

cussed here, but may be found in the original report. Never-

theless, some important points are worth mentioning. In this

model, water vapor saturation of the air stream is not a basic

assumption as was the case for the heat transfer model develop-
ed for the mechanical draft tower. Instead, the sensible heat

transfer is calculated in addition to the total heat transfer

due to both evaporation and sensible heat transfer. As in

the mechanical draft tower model the transfer calculation is

based on a finite-difference approximation to the Merkel

Equation, but in this case the counter-flow of the air and

water streams necessitates only a one-dimensional calculation.

32

Fig. 2.3
Comparison of Reported and Predicted Mechanical Draft
Coolina Tower Performance

Water Loading = 6200 ibm/hr - f t 2

Air Loading = 1692 lbm/hr - ft2

95

90
Tower
Discharge
Temp.(OF)
85

- reported
80
0 -predicted

75

55 75
Wet Bulb Temoerature (OF)
33

The calculation of both the total energy transfer and.

the sensible heat t-ransfer allows the determination of the

exact psychrometric condition (both dry bulb and humidity)
of the air stream leaving each "cell" of the finite differ-

ence integration. The assumption of water vapor saturation,

if it in fact did not exist, would result in an underesti-

mate of the fill air exhaust dry bulb temperature and hence
an underestimate of the induced draft.
To complete the thermal model of a natural draft tower
a relationship between the tower height and the tower base
diameter needed to be established for different sized towers.
This was necessary because while a mechanical draft tower
may be sized to a particular cooling duty.by varying the

number of tower modules, a natural draft tower is sized by

varying the tower size. Flangan [7] has published data

concerningthe ratio of height to diameter for 16 large natur-

al draft towers which indicates an average ratio of 1.248.

2.5 Cooling Ponds

2.5.1* Literature Review

The task of mathematically modeling the thermal-hydraulic

behavior of a cooling pond is a problem which is substantially

different from the problem of modeling cooling towers. This

is because actual cooling ponds are not physically well-defined
in the sense that the important parameters which determine
34

their thermal behavior can not be assigned values which are

representative of all, or even most. cooling ponds. In

fact different cooling ponds may exhibit completely differ-

ent types of thermal-hydraulic behavior each of which require

different analysis approaches and techniques.

There are two idealized cases of pond thermal-hydraulic

behavior which yield themselves to very simple analytical

treatment [8],. These are termed the plug-flow and fully-

mixed models. In plug flow there is no mixing between the

discharge into the pond and the receiving water and the sur-
face temperature, for steady-state conditions, decreases
exponentiall-y from the pond inlet to the pond outlet. The

fully-mixed pond represents an extremely high degree of mix-

ing of the discharge and the receiving water. Thus a uni-
form temperature over the entire pond results. In reality,
the behavior of most ponds would fall between these two

extreme cases. The plug flow pond represents the best possi-
ble heat dissipation situation since the temperature of the

discharge is kept as high as possible. Conversely, the

fully-mixed pond represents a lower bound on the heat trans-

fer performance of the pond. The "worst case" performance,
however, is a short-circuited pond. For either the plug-flow
or fully-mixed model both steady-state and transient behavior
can be readily calculated.
35

Ryan [5] reported the development of a transient

cooling pond thermal-hydraulic model which was the first

attempt to realistically mathematically model the actual
physical process occurring in a cooling pond. Watanabe [6]
extended the model and reported criteria for its applicabi-

lity. This model is recommended for use as a design tool or

means of evaluating the performance of cooling ponds rela-
tive to alternative waste heat disposal systems. However,
since the model is not fully developed into a documented
computer program its applicaetion appears difficult. Also,
for the purposes of most surveys the computational time is
excessive.

2.5.2 Selection of a Model

The task of formulating a representative thermal-hydrau-
lic model of a cooling pond can be considered to be differ-

ent from the task of formulating a model of a cooling pond

which is to be used for design purposes. The present inter-
est is in mathematically representing the approximate thermal-
hydraulic behavior of a representative cooling pond. It is
perceived that this limited goal can be accomplished through
the use of a plug-flow, vertically-mixed pond model capable

of accounting for variable meterological conditions, variable

inlet temperatures, and variable flow rates. For a given
36

cooling requirement such a model would tend to predict pond

sizes which are smaller than would be normally required.

Thus, if the model were to be used in a detailed economic
comparison of alternative waste heat disposal systems the
pond economics would be unduly favored.
The vertically-mixed, plug-flow model predicts the
transient pond behavior by following a slug of water of uni-
form temperature through the pond and calculating the aver-
age heat loss for each successive day of residence in the
pond. The heat transfer correlations used in this model are

those recommended by Ryan [5] . The basic equation of the

net energy flux from a water surface exposed to the environ-

ment is

$n r - [4. Ox10~ 8 (Ts+460) + FW[ (es-ea) + 0.25(T-Ta]

(2.20)

where FW = 17*W for an unheated water surface,

FW = 22.4(Ae) 1 / 3 + 14*W,

AO = Tsv - Tav (*F),

W = wind speed at 2 meters (MPH),

Tsv = virtual temperature of a thin vapor layer in
contact with the water surface,
= (TS + 460)/(l - .378 es/P),

Tav = virtual air temperature,

= (Ta + 460)/(l - .378 ea/P),

37

es= saturated vapor pressure at T (mm Hg),

ea = satu.rated vapor pressure at Ta (mm Hg),

P = atmospheric pressure (mm Hg),

Ts= bulk -water surface temperature (*F),

Ta air dry. bulb temperature (*F),

2
On = net heat from pond surface (BTU/day-ft
$r = Osn + $an = net absorbed radiative energy,
$sn = net absorbed solar radiation,
= .94($sC)(1 - 0.64C2

$sC = incident solar radiation,

C = fraction of sky covered by clouds,

. = net absorbed longwave radiation, and

an
= 1.16x10- 1 3 (460 + Ta) 6 (1 + 0.17C2).

The computational algorithm for the plug-flow model is

given in Fig. 2.7. Note that the model is not a perfect plug-
flow model in that each plug of water entering the pond is

assumed to be mixed with the slug immediately preceeding it.

This mixing qualitatively accounts for the effect of entrance

mixing.

2.6 Dry Cooling Towers

In relation to the other waste heat dissipation systems,

the development of a reliable performance model of dry cooling

towers is simple. The amount of heat rejected by a mechanical

draft dry tower can be shown to be directly proportional to
38

Fig. 2.7

Cooling Pond -Model Computational Algorithm

Assign to each discharge slug an identifi-

cation number = day. of discharge

.I initialize slugs in pond at time zero; temp-

erature, volume, fraction of volume in pond

begin calculation for day J

calculate temperature change for all slugs

in pond

add discharge volume of day J to pond volume,

sdt fraction of volume of slug J in pond equal
to 1

mix slugs J and J-1

determine which slugs (and fractions thereof)

remain in pond by summing up slug volumes for
day J, J-1, J-2, ... ; until pond volume is
equaled or exceeded

determine which slugs (and fractions thereof)

exhausted from pond during day J by comparing
new pond inventory (day J) and old pond
inventory (day J-1)

mix all exhausted slugs to find withdrawal

temperature for day J

end day J calculation

39

the difference between the inlet water temperature and inlet

air temperature for a fixed dry tower design. With refer-

ence to Fig. 2.8

Q =UAAT 1m F (2.21)
g

where Q = heat rejection rate,

A = heat transfer surface area,

U = effective heat transfer coefficient,

F = cross-flow correction factor, and

g
ATlm= log mean temperature difference.

(T0 -Ti) - (T-T )

ATm T -T ) (2.22)
In 0i

where (T9-T') > (T -T')

T = water inlet temperature,

Ti = water outlet temperature,

T = air inlet temperature, and

i
T = air outlet temperature.

A heat balance on the tower gives

LCw(Ti-To ) = GC a (T9-TI)
o i (2.23)

Equations (2.21), (2.22) and (2.23) may be combined to

yield
40

Fig. 2.8
Dry Tower Schematic Drawing

L T
G

Water Air
Flow Flow

L G5T
41

Q ITD(ex) (2.24)
e 1
GCa LCw

where ITD= T -T ,

and

x = F UA -

Now note that, for fixed values of the parameters U,

A, Fg, G and L,

Q a ITD (2.26)

This result has been found by Rossie [1] to be experi-

mentally verified. Further, Rossie has found that the ther-
mal performance of natural draft dry cooling towers may be

reasonably expressed by a relationship of the form

Q a ITDb (2.27)

where b is a constant for a given tower. A typical value of

b is 1.33.
42

REFERENCES

1.0 Rossie, J.P., "Research on Dry-Type Cooling Towers

for Thermal Electric Generation, Part 1," Water
Pollution Control Research Series, EPA, 16130EE511/70.

2.0 Croley, T.E., "The Water and Total Optimizations of

Wet and Dry-Wet Cooling Towers for Electric Power
Stations, Iowa Institute of Hydraulic Research Report
#163, Jan. 1975.

3.0 "Heat Sink Design and Cost Study for Fossil and Nuclear
Power Plants," WASH-1360, USA AEC, December 1974.

4.0 "Kool-Flow Spray Cooling Modules," Richards of Rock-

ford Technical Manual.

5.0 Ryan, P.J., Harleman, D., "An Analytical and Experi-

mental Study of Transient Cooling Pond Behavior," MIT
Ralph M. Parsons Lab. Report #161, Jan. 1973.

6.0 Watanabe, M., Harleman, D., "Finite Element Model of

Transient Two Layer Cooling Pond Behavior," MIT
Ralph M. Parsons Lab. Report #202.

7.0 Flanagan, T.J., MIT Master's Thesis, 1972.

8.0 "An Engineering-Economic Study of Cooling Pond Perfor-
mance," Littleton Research and Engineering Corporation,
Water Pollution Control Research Series, EPA,
16130DFX05/70.

9.0 "Survey of Alternate Methods for Cooling Condensor

Discharge Water," Dynatech R/D Company, Water Pollution
Control Reserach Series, EPA, 16130DHS11/70.

10.0 Shiers, P.F., Marks, D.H., "Thermal Pollution Abatement

Evaluation Model for Power Plant Siting," MIT-EL-73-013
Feb. 1973.

11.0 Merkel, F., "Verdunstungskulung," VDI Forschungsar-

beiten, No. 275, Berlin, 1925.

12.0 Hallet, G.F., "Performance Curves for Mechanical Draft

Cooling Towers," ASME Paper #74-WA/PTC-3.

13.0 Nahavandi, A.N., "The Effect of Evaporative Losses in

the Analysis of Counter-flow Cooling Towers," Unpublished
Paper, Newark College of Engineering, Apr. 1974.
43

14.0 Yadigaroglu, G., Pastor, E.J., "An Investigation of

The Accuracy of the Merkel Equation for Evaporative
Cooling Tower Calculation," ASME paper #74-HT-59,
AIAA Paper #74-765.

15.0 Marks, "Handbook of Mechanical Engineering," Chapter

15, McGraw-Hill, 1968.

16.0 Cooling Tower Institute Cooling Tower Performance

Curves, 1967.

17.0 Hoffman, D.P., "Spray Cooling for Power Plants,"

Proceedings of the American Power Conference, Vol. 35,
1973.
18.0 Porter, R.W., "Analytical Solution for Spray Canal
Heat and Mass Transfer," ASME paper 74-HT-58, AIAA
paper 74-764, July, 1974.

19.0 Porter, R.W., Chen, K.H., "Heat and Mass Transfer in

Spray Canals," ASME paper 74-HT-AA, Dec. 1973.

20.0 Porter, R.W., Personal Communication, June 1975.

21.0 Chilton, H., "Performance of Natural-Draft-Water-Cool-

ing Towers, Proc. IEE, 99. pt. 2, pp. 440-456, 1952,
London.

22.0 Keyes, R.E., "Methods of Calculation for Natural Draft

Cooling Towers," HEDL-SA-327.

23.0 Lowe, H.J., and Christie, "Heat Transfer and Pressure

Drop Data on Cooling Tower Packing, and Model Studies
of the Resistance of Natural Draught Towers to Airflow,"
International Heat Transfer Conference, Denver, 1962,
pp. 933-950.
24.0 Winiarski, L.D., "A Method for Predicting the Performance
of Natural Draft Cooling Towers," EPA, Thermal Pollution
Research Program, Report #16130GKF12/70.

25.0 Rish, R.F., "The Design of Natural Draught Cooling

Tower," International Heat Transfer Conference, 1962
Denver, pp. 951-958.

26.0 Kennedy, John F., "Wet Cooling Towers," MIT Summer

Course on the Engineering Aspects of Thermal Pollution,
1972.
27.0 McKelvey, K.K., Brooke, M., "The Industrial Cooling
Tower," Elsivier Company, Amsterdam, 1958.
28.0 Guyer, E.C., Sc.D. thesis, "Engineering and Economic
Evaluation of Some Mixed-Mode Waste Heat Rejection
Systems for Central Power Stations," MIT, 1976.
45

APPENDIX A

COMPUTER PROGRAMS FOR WASTE HEAT REJECTION

SYSTEM THERMAL PERFORMANCE CALCULATIONS

This Appendix contains FORTRAN computer programs for

predicting the thermal performance of a mechanical-drfat

evaporative cooling tower (MECDRAFT), a natural-draft evap-

orative cooling tower (NATDRAFT), and a spray canal (SPRANAL).

Programming of the cooling pond and dry cooling tower models

discussed in this report may be easily done by the user.

Tables A.l to A.3 list the required input variables for the

three programs. The FORMAT of the required input may be

easily obtained by examining the appropriate program listing.

46

Table A.l

Required Data Input for Natural Draft Evaporative

Cooling Tower Model (NATDRAFT)

Definitions of Input Variables in Order of Occurrence

WTRF = Water flow rate (lbm/ft 2 -hr)

HUM = Relative humidity of ambient air (%/100)

AIRTI = Inlet air temperature (*F)

WTRTI = Inlet water temperature (OF)

WTRTOA = Initial geuss at water outlet temperature (OF)

Other Variables Defined Internally in the Porgram Which May

Be Adjusted

AIRF = Initial geuss at air flow rate (lbm/hr-ft2 )

PPP = Type of Tower Packing

HPACK = Height of Tower Packing (ft)

HAIRIN = Height of Tower Packing Above Ground (ft)

ATOTAL = Total Packing Surface Area per unit flow area

ADPK = Surface Area per unit flow area for computing 2

pressure loss in packing due to skin friction (ft )

AFPK = Fraction of Tower cross-section which is unobstruc-

ted by packing

HTOWER = Height of Tower Chimney (ft)

DTOWER = Diameter of Tower at Base (ft)

47

Table A.2

Required Data Input for Spray Canal Model

(SPRANAL)

All Required Data to be Defined Internally in Program

R = Fraction of total water flow sprayed by each

spray device

TEMDIS = Canal inlet water temperature (OF)

TWB = Ambient wet-bulb temperature (OF)

WSPEED = Ambient wind speed (MPH)

PASSES = Number of spray passes marching down canal

NROW = Number of rows of spray devices across canal

48

Table A.3
Required Data Input for Mechanical-Draft
Evaporative Cooling Tower Model
(MECDRAFT)

Program presented as a complete SUBROUTINE TOWER(J) which

determines the tower output temperature CTWOUT on Day J
with an ambient wet bulb temperature TWBXX(J). All data
input through COMMON statements.

TTOWIN = tower inlet water temperature (*F)

PSA = vapor pressure of water at 1 OF increments
from 0 to 150 OF (psia)

WATERL = water flow per square foot of tower cross-

section (lbm/hr-ft 2 )
WL = total water flow to tower (lbm/hr)

Other variables defined internally in program which may be

adjusted.

ACELLW = calculational cell water loading area (ft 2

ACELLA = calculational cell air loading area (ft2)
N = square root of number of calculational cells
HEIGHT = height of tower fill (ft)

--
C PROGRAM FOR PREDICTING THE THERNAL PERFORMANCE OF A NATURAL DRAFT
C COOLING TOWER - NATDRAFT--
C ='.A N. IZED CROSS-SECTION.L DR AREA - -IR I :T
C ADOT=NORBALIZED CROSS-SECTIONAL DRAG AREA AT AIR OUTLET
C ADPK=SURFACE AREA PER UNIT FLOW ASA FOR COMPUTING PRESSSUE LOSS IN
C PACKING DUE TO SKIN FRICTION LOSS
C ADSL=NORNALIZED CROSS-SECTIONAL AREA FOR DRAG IN SHELL
C AFIN=NORNALIZED CROSS-SECTIONAL FLOW THROUGH AREA AT THE All INLET
C AFPK=PORTION OF TOWER-SECTION WHICH IS UNOBSTRUCTED BY PACKING
C AFOT=NORMALIZED CROSS-SECTIONAL FLOW AREA THROUGH OUTLET OF PACKING
C AFSL=NORNALIZED CROSS-SECTIONAL FLOW THROUGH AREA IN THE SHELL
C AIRF=INITIAL GEUSS FOR THE AIR FLOW RATE
C AIRTI=INLET AIR DRY BULB TEMPERATURE
C ATMOS=ATMOSPHERIC PRESSUR!
C ATOTAL=TOTAL PACKING SURFACE AREA IN ONE SQUARE FOOT OF TOWER X-SECTION
C CDIN=DRAG COEFFICEINT FOR INLET STRUCTURE
C CDOT=DRAG COEFFICIENT FOR CUTLET STRUCTURE
C CDSL=DRAG COEFFICIENT FOR THE SHELL
C CP=SPECIFIC HEAT OF AIR
C DTOVER=TOWER DIAMETER AT PACKING
C HAIRIN=HEIGHT OF PACEING AIR INLET
C BPACK=HEIGHT OF THE PACKING
C HTOVER=TOVER HEIGHT
C HUM=RELATIVE HUMIDITY OF INLET AIR
C LANBDA=EMPIRICAL COEFFICIENT FOR SPLASH PACKING
C N=EMPIRICAL COEFFICIENT FOR SPLASH PACKING
C P1,,P16,P23,P26=EMPIRICAL PRESSURE DROP DATA---LOWE AND CHRISTIE
C SPACE=CENTER TO CENTER SPICING OF PARALLEL PLATES
C THICK=THICKNESS OF PARALLEL PLATE.PACKING
C T OLERR=COTIERG ENCE TOLER-A*Cl FOR TOW IEIGHT
C TOLERT=CONVERGENCE TOLERANCE FOR INLET WATER TEMP
C WTRF=NORMALIZED WATER FLOW RATE
C WTRTI=INLET WATER TEMPERATURE
C WTRFT=TOTAL WATER FLOW RATE
C WTRTO=INITIAL GEUSS FOR OUTLET WATER TEMP
C A=INTEGRATION SEGMENT AREA
- -1-,,-,-- -4-

C AIRFL-CURRENT VALUE OF AIR FLOW RATE

C AIRT=CURRENT AIR TEMPERATURE
C C=TEMPORABY VARIALBE
C CF=FRICTION COEFFICIENT
C CONTR=VEIGTH OF CONDENSED WATER
C DA=&REA SEGMENT
C DAIRT=CHANGE IN AIR TEMPERATURE DURING ONE INTEGRATION STEP
C DNSARI=DENSITY OF INLET AIR
C DNSAVG=AVERAGE AIR DENSITY
C DNSARO= DENSITY OF OUTLET AIR
C DTODTI=R&TE OF OULET WATER TEMP CHANGE VERSUS INLET WATER TEMP CHANGE
C DWTRT=CHANGE IN WATER TEMPERATURE DURING ONE INTEGRATION STEP
C ENT=AIR ENTHALPY IS INTEGRATION PROCEEDS
C ENTI=ENTHALPY OF INLET AIN
C ENTSA=ENTHALPY OF AIR DURING THE SATURATION ADJUSTMENT LOOP
C ENTSAT=ENTRALPY OF A PCUND OF SATURATED AIR-WATER MIXTURE
C H=CALCULATED TOWER HEIGHT
C H1,H2=HOLDING VALUES OF TOWER HEIGHT
C HENT=ADJUSTED ENTHALPY OF AIR-WATER DROPLET MIXTURE IN SATURATION
C ADJUSTMENT LOOP
C HG=HEAT TRANSFER COEFFICIINT
C HUNI=RELATIVE HUMIDITY AS INTEGRATION PROCEEDS
C LBW=POUNDS OF WATER PER POUND OF AIR AT ANY POINT IN PACKING
C LBVI=POUNDS OF VAPOR PER POUND OF AIR
C LBVLBA=POUNDS OF VAPOR PER POUND OF AIR AT ANY POINT IN THE PACKING
C NOITER=NURBER OF ITERATIONS COMPLETED
C PRLIW=PRESSURE LOSS AT THE INLET
C PRLPK=PRESSURE LOSS IN PACKING
C PRLPR=PRESSURE LOSS DUE TO PROFILE
C PRLOT=PRESSURE LOSS AT OUTLET
C PRLSL=PRESSURE LOSS IN SHELL
C PRLSP=PRESSURE LOSS DUE TO SPRAY
C PSA=SATURATION VAPOR PRESSURE AT THE AIR TEMPERATURE
C PSAH=SATURATION VAPOR IN SATURATION ADJUSTEENT LOOP
C PSATO=SATURATION VAPOR PRESSURE
C PSV=SATURATION VAPOR PRESSURE AT TUE WATER TENEfRATURE
C VHSP=VELOCITY HEADS LOST DUE TO SPRAY
C VHVC=VELOCITY HEADS LOST DUE TO TENA-CONTRACTA IN THE TOWER
C VIN=AIR INLET VELOCITY
C VNON=NONINAL VELOCITY IN PACKING
C VPEN ENTHALPY OF MOISTURE IN AIR, USED IN SATURATION ADJUSTMENT LOOP
C VPENT=ENTHALPY OF VAPOR IN AIR
C VPRES=VAPOR PRESSURE OF AIR
C VPK=AIR VELOCITY IN PACKING
C VOT=AIR VELOCITY AT OUTLET
C VSL=AIR VELOCITY IN THE SHELL
C WTRLT=VATER WHICH COWDENSES OUT DDRI5G N INTEGRATION STEP
C WTRT1,VTRT2=HOLDS WATER INLET TEMPERATURE FOR EXTRAPOLATION
C LOGICAL VARIABLES
C ENDFLG= TRUE IF PROGRAM HAS REACHED NORMAL TERMINATION
C EXTIFL=TRUE IF ITERATION IS BEING HADE TO EXTRAPOLATE AIRFLOW
C EITTO=TRUE IF ITERATION IS BEING MADE TO EXTRAPOLATE OUTLET WATER TEMP
C PPP=TRUE IF TOWER HAS PARALLEL PLATE PACKING
C PRII=TRUE IF SPLASH PACKING
C PRITER=TRUE IF RESULTS OF EACH ITERATION ARE TO BE PRINTED
C PRSTEP=TRUE IF EACH STEP IN ITERATION IS TO BE PRINTED
LOGICAL ENDFLG, PRITER, PRSTEP, EXTITO, EXTAFL
LOGICAL PPP
REAL LBVLBALBWLANEDAN,LBVLBSLBVIKAL
1111 CONTINUE
READ(5,999) VTRF,HUMAIRTI,WTRTIWTRTOA
999 FORMAT(5F10.3)
FACTOR=3.2
1001 CONTINUE
VTRTO=WTRTOA
AIRF=1264.0
C**** ***************************************************************************
C IMPORTANT### SET PPP TRUE IF PARRALLEL PLATE PACKING IS USED
C IF PPPP IS TRUE RISH'S HT TRANSFER AND PRESSURE DROP REALATIONS
C ARE USED
C*******************************************************************************
PPP=. TRUE.
I-b
RPACK=5.33
HAIRIN=35.6

C SKIP THE NEXT INPUT PARANETERS IF PARRALLEL PLATE PACKING IS USED

IF(PPP) GO TO 2
LANBDA=0.065
N=.6
P 13=1.2
P16=0.9
P23=2.0
P26=1.3
ATOTAL=HPACK

2 CONTINUE

C IF PPP IS TRUE INPUT ATWCER AND APPK AS REQUIRED

C *******************************************************************************
IF(.NOT.PPP) GO TO 7
ATOTAL=204.6
ADPK=204.0
AFPK=.70
7 CONTINUE
CP=0.24
ATROS=14.4
AFIN=1.0
AFOT=1.0
AFSL=1.0
ADIW=0.0
ADOT=0.0
ADSL=0.0
CDOT=0.0
CDSL=0.0
CDIN=0.0
TOLERT=0.3
TOLERH=10.0
HTOVEWR=51 4.0
DTOVER=372.0
STEPS=2Q .0
LS?1P-50
LP§DFLG *FALSE.
PRITRR=.TRE.
PESTEP=. FALSE.
EXT WC *FALSE.
EXT &FL-=.FALS2.
L1TER=52
-AIR?=AIR?!
W0ITh'=0
VHYC0. 167* (DTOVER/HAI.RlU)**2
VPRES=HUM*PSA? (AIR?)
LBVLBA=0 .622*VPRES/ (AIMOS-VPRES.)
VPZNT=l06l1.04. 44*AIRT
IWI=CP* (AIRT- 32.0) *VrEI1*LBVLE+A
VPR9S I=VPRES
LBVI=LBVLBA
DWSARI= ( (ATMOS- VPRES)/53. 3 VPPlS/85 7)+*1414wO/ (#60.-04&IiT)
DA=ATOTAL/STBPS
AIRFL=0,0

C
C END INPUT! AND INITIALIZATION
C START ITERATION
C

95 VNOB=AIRP/ (DNSARI*3600.0)
VHSP=0. 16*HRIRIV* (WTRP/AIRF) **1 .32
IF-(PPP) GO TO 16
K&L-HPACK*LAIBDA* (AIRI/VTRF) **Jf
GCP*WTRF*KAL/HPACK
HGOUT=0 .0
Tl=VN0t /3.0-1 .0
Pl-(P16-PI3) *Tl+P1.3
P2= (P26-P23) *T14.P23
VHLPK=((P2-P1)*(WTRF-1000.O)/1000.O+P1)*HPACK
cp=oo*
GO TO 15
16 CP=O.0 192* jVTRP/AIRF) **O.5
CF=PACTOR *CF
HG=CP*AIR*CF/(2.OtCP*71,6* (AIRP/VTRF) **O.25)
K&AL5.G*ATOTAL/ (CP*WTRY)
HGOUT=HG
15 NTRTM-TRTO
Eu'r=ENTI
HUftI=Holl
A=0.0
LBVLBA=LBVI
VPRBS=VPRES I
COweri=o. 0
AIRT= AIBTI

c INTZGRATION LOOP BEGINS VITH STAT~EENT 6

6 PSW=PSAT(VTRT)
IV(PSW.BQ.O.0) GO TO 110
RWTSAT=CP* (WTRT-32.O) +(1061.O+.44#*1RT):*O,.622*PSV/(ATNOS-PSV)
C=HG*Dk* (ENTS AT-ENT) /CP
IF(. NOT. PlSTEP.OR.lXTVTO.OR.E?!AFL) GO TO .35
IP(LSTIPLT.47) GO T.9 ~

37 PORSAT(52H1COOLING, TOVEtt PROGRAM E STEP RESULTS-OF ONEw

BYTE
1 10Hi ITERATION/57H0 V&TER AIR SATUE ACTUAL EEL PNDS V
2TR/ VAPOR/56f AREA TYEMP TEMP ENTHAL ENTHAL HUN PIDS AIR
3PRES)
LSTZP=0
LITER=52
36 LSTEP=LSTEP+1
WRITE(6,38) AuWTRTiAIRTENTISATuENTgHUNIuLBVLBAerVPRES
38 FORMAT (5F7. I,F6. 3 ,F9 *5 vF7 * 4)
35 DVTRT=C/VTRF 4r
DENT=C/AI RF
DAIRT=HG*Dk* (WTRT- AIRT) / (AIRF*CP)
VTRT=WTR T+DWTRT
ENT=ENT+DENT
AIRT=AIRT+DAIRT
A=A+DA
VPENT=1061.0+0.444*AIET
LBVLB A=(ENT-CP* (AIRT-32.0)) /VPENT
PSA=PSAT (AIRT)
IF(PSA.BQ.0.0) GO TO 110
LBVLBS=0.622*PSA/(ATNOS-PSA)
HUNI=LBVLBA*(0.622+LBVLBS)/(LBVLBS*(.622+LBVLBA))
VPRES=HUI*PSA ,
IF(HUNI.LE.1.0) GO TO 99

C IF MIXTURE IS SUPER-SATURATED, FIX-UP

C**** ******** ******************* ***** *******************************************
T= AIRT
97 T=T+O.1
PSAH=PSAT(T)
IF(PSAH.EQ.0.0) GO TO 110
VPEN=1061.0+.444*T
LBW=0.622*PSAH/( TMOS-PSAH)
ENTSA=CP* (T-32.0) +VPEN*LBW
HENT= (LBVLBA-LBV+CONWTR) * (T-32.0) *ENTSA
IF(ENT.GT.HENT) GO TO 97
CONWTR=LBVLBA-LBV+CONWTR
ENT=ENTSA
AIRT=?
99 IF(A.LT.ATOTAL) GO TO 6

C END INTEGRATION SECTION

C
C COMPUTE PRESSURE LOSSES FOR THIS ITERATION

100 IF(EXTVWTO) GO TO 24.

YPBNT= 1061.0+0.444*AIRT
LBVLBA= (ENT-CP* (AIRT- 32.0)) /VPENT
VTRLT=AIRF* (LBTLBA+CONVTR -LBVI)
VPRES=LBVLBA*ATNOS/(0.622+LBVLBA)
DWSARO= ((ATMOS-VPRES)/53.3+VPRES/85.7) *144.0/(460.0+IRT)
DNSARO=DNSARO*(1.O+CONTR)/(1.0+CONVTR*DNSARO/62.4)
DNSAVG= (DNSAIRI+DNSARO)/2. 0
VIN=VNOM/AFIN
VOT=AIRP/(DNSARO*APOT*3600.0)
VSL=AIRF/(DNSARO*APSL*3600.0)
PRLIN=CDI N*DNSARI*0 .0 16126*ADIN*VIN**2
IF(.NOT.PPP) GO TO 102
VPK=AIRP/(DNSAVG*AFPK*3600.0)
PRLPK=CF*DNSAVG*0.016126*ADPK*VPK**2
GO TO 103
102 PRLPK=DNSARI*0.016126*VHLPK*VNO0**2
VPK=VNOM
103 PRLOT=CDOT*DNS&RO*0.0 16126*ADOT*VOT**2
PRLSL=CDSL*DNS ARO*0.0 16126*ADSL*VSL**2
PRLVC=VHVC*DNSARI*0.0 16126*VNOH**2
PRLSP=VHSP*DNSARI*0.O 16126*VNON*VNON
PR LPR=PRLOT+PRLIN+PRLSL
H=(PRLPR+PRLPK+PRLSP+PRLVC)/ (DNSARI-DNSARO)
IF(ENDYLG) GO TO 40
NOITER=NOITER+1
IP(.NOT.PRITER.OR.EXTAL) GO TO 21
40 IF(LITER.LT.52) GO TO 30
LSTEP=50
LITER=O
WRITE (6,31)
31 FORMAT(46H1COOLING TOVER PROGRAM - RESULTS OF ITER&TIONS/
I 221, 17HAIR CALC TOWER/
263H OUTLET VELOTY HEAT CHARAC- SKIN INLET
3, 50H OUTLET OUTLET PROFILE PACKING SPRAY VENA CON/
463H ITER WATER AIR IN TRANS TERISTIC FRICTION RELAT WATER
5, 56H AIR AIR PRESSURE PRESSURE PRESSURE PRESSURE TOWER/
663H NO LOSS DENSITY PAKING COEFF (K*A/L) COEFF HUMID TEMP
7, 57H TEMP ENTHAL LOSS LOSS LOSS LOSS HEIGHT)
30 WRITE(6,32)NOITER,WTRLTDNSAROVPKRGOUTKALCF,HURI,WTRTAIRT,
1 ENTPRLPRPRLPK,PRLSPPRLYC,H
32 FORMAT(1HOI4,F7.2,F8.6,F7.3,F6.3,F8.4,F9.5,77.3,F6. 1,
1 F6.1,F7.1,F0.6,3F9.6,F7.0)
LITER=LITER+2
IF(ENDFIG) GO TO 33

C END PRINTING RESULTS OF ONE ITERATION

21 IF(NOITER.LE.100) GO -tO 39
WRITE (6,98)
98 FORMAT(47H MORE THAN 100 ITERATIONS. EXECUTION TERMINATED)
GO TO 39
C ITERATION STOP BYPASSED
STOP

C NOW FIND IF SPECIFIED TOLERANCES ARE NET

39 IF (ABS (WTRT-WTRTI) .LE.TOLERT) GO TO 27

IF(.NOT.PRITER) GO TO 46
IF(.NOT.EXTAFL) GO TO 48
WRITE (6,42)WTRTO
42 FORMAT(30H (EXTRAPOLATING FROM VTRTO=,F. 1, iH))
LITER=LITER+1
GO TO 46
48 VRITE(6,43) WTRTO
LITER=LITER+2
43 FORNAT(31H (EXTRAFOLATING FROM VTRT0=,F6.1,1H))
46 WTRT1=UTRT
WTRTO=WTRTO+O.1
EXTVTO=.TRUE.
GO TO 15
27 IF (EX TAFL) GO TO 50
IF(ABS(H-HTOWER).LE.TCLERH) GO TO 29
IF(.NOT.PRITER) GO TO 44
WRITE(6,41) AIRF
LITE=LITER+2
41 FORMAT(26H0(EXTRAPOLATING FROM AIRF=,F7.1rlH))
44 AIRFL=AIRF
H 1-=R
AIR?= AIR+ 10. 0
EXTAFL=.TRUE.
GO TO 95

C A SAMPLE ITERATION HAS BEEN MADE TO ADJUST AIRF OR VTRTO

50 H2=H
DAFDH=10.0/(H2-H1)
EXTAFL=.FALSE.
AIRF=AIRF+DAFDH* (HTOIR-5)
I?(.WOT.PRITER) GO TO 95
WRITE (6,55) AIRF
LITER=LITER+ I
55 FORMAT(20H (MODIFYING AIRF TO ,?7.1,1H))
GO TO 95
24 VTRT2=WTRT
DTODTI=0. 1/(WTRT2-WTRT 1)
EXTVTO=.FALSE.
UTRTO1=VTRTO
WTRTO=VTRTO+DTODTI*(INYRTI-WTRT)
IF(.NOT.PRITER) GO TO 15
IF(.WOT.EXTAFL) GO TO 62
WRITE(6,61) WTRTO
61 FORMAT(25H (MODIFYING RTRTO TO P?6.1t18))
LITER=LITER+1
GO TO 15
62 WRITE (6,60)WTRTO
LITER=LIT ER+2
60 FOREAT(21H (MODIFYING UTRTO TO ,F6.1,1R))
GO TO 15
29 IF(PRITER) GO TO 33
ENDFLG=.TBUE.
LITER=52
GO TO 100
33 WRITE(6,96)UTRTO, H
96 FORNAT(26H END COOLING TOWER PROGRAN/34HOFINAL OUTLET WATER TEMP
1ERATURE ISF6.1/22HOFINAL TOWER HEIGHT IS,F7.0)
WRITE(6,1002) FACTOR
1002 FORNAT(8H FACTOR=,F10.4)
RAVGE=WTRTI-W TRTO
WRITE (6, 998)
998 FORMAT(102H WATER LOAtING HUMIDITY INLET AIR TEMP INLE
IT WATER TEMP ACTUAL OUTLET UATER TEMP RANGE)
WRITE (6,997)WTRFRUNAIRTI,NTRTI, WTRTOA, RANGE
997 FORMAT (2XF1O.3,6XF1O.3,6XF10.3,8,F1O.3,13XF1O.3, 1,F1O.3)
GO TO 1111
STOP
110 AIRF= (AIRF-AIRFL)/2.04&IRFL
IF(.NOT.PRITER) GO TO 95
WRITE(6,111) AIRF
LITER=LITER+2
111 FORMAT (19HO(ADJUSTING AIRF TO,7.1, 15H FOR STABILITY))
GO TO 95
END
FUNCTION PSAT(T)
DIMENSION V(181)
DATA M/0/
D ATAV/.08854,. 09223,.09603,.09995,.10401, .10821,.11256,. 11705,. 121
170,. 12652,. 13150,. 13665,. 14 199,. 14752,. 15323,. 159 14,. 16525,. 17157,
2.17811,.18486,.19182, .19900,.20642,.2141, .2220,.2302,.2386,. 2473,.
32563,.2655,.2751,.2850,.2951,.3056,.3164,.3276,.3390,.3509,.3631.
43756,.3886,.4019, .4156,.14298,.4443,. 4593 .747,.4906,.5069,.5237,.
55410,.5588,.5771,.5959,.6152,. 6351,.6556,. 6766,.6982,.7204,.7432,.
67666,.7906,.8153,.8407,.8668,.8935,.9210,.9492,.9781,1.0078,1.0382
7, 1.0695, 1. 10 16, 1. 1345, 1. 1683, 1.2029,1.2384,1.2748, 1.3 121, 1.3504, 1.
83896, 1. 4298, 1.4709, 1.5130, 1.5563, 1.6006, 1.6459, 1.6924, 1.7400, 1.788
98, 1.8387, 1.8897, 1.9420, 1.9955,2.0503, 2.1064,2.1638,2.2225, 2.2826, 2
*.3440,2.4069,2.4712,2.5370,2.6042,2.6729,2.7432,2.8151,2.8886,2.96
*37,3.0404,3.*1188,3. 1990,3.281,3.365,3.450,3.537,3.627,3.718, 3.811,
*3.906,4.003,4.102,4.203,4.306,4.411,4.519,4.629,4.741,4.855,4.971,
*5.090,5.212,5.335,5.46l,5.590,5.721,5.855,5.992,6.131,6.273,6.471,
*6.565,6.715,6.868,7.024,7.183,7.345,7.510,7.678,7.850,8.024,8. 202,
*8.383,8.567,8.755,8.946,9.141,9.339,9.541,9.746,9.955, 1O168,10.38
*S,10.605,1O.830,11.058,11.290,11.526,11.769,12.011,12.262,12.512,1
*.771,13.031, 13.300, 13.568, 13.845, 14. 123, 14.4110, 14.696/
NT=T
PSAT=O.O
IF(NT.GT.31) GO TO 5
PSAT=V(1)
WRITE(6,2) T
2 FORIAT(36HOERROR IN PSAT: TABLE EXCEEDED. T=,?8.2)
4 M=N+1
IF(.LB.50) RETURN
WRITE (6,3)
3 FORNAT(53H0 MORE THAN 50 ERRORS IN PSAT - EXECUTION TERMINATED)
STOP
5 IF (NT.GE.212) GO TO 4
1 PSAT=V(NT-3 )+(Y (WT-30)-V(NT-31))*(?-NT)
RETURN
END
C PROGRAM FOR PREDICTING THE THERMAL PERFORM NCE 0 SPRA SPRANA
C E=TOTAL WATER EVAPORATED
C ALPHA=FRACTION OF WATER EVAPORATED
C F=AIR INTERFERENCE FACTORS
C PSA=SATURATION VAPOR PRESSURE
C TBIX=MIXED CANAL TEMPERATURE
C TST=SPRAY TEMPERATURE
C TVBL=LOCAL WET BULB TERPERATURE
C PASSES=NUMBER OF PASSES MARCHING DOWN CANAL
C NRON=NUMBER OF SPRAYS ACROSS CANAL
C R=FRACTION OF WATER SPRAYED BY EACH SPRAY 0
C TEMDIS= CANAL INLET TEMPERATURE
SSVd, 03VaI HiOJ KOIJlflf!31VD liISIR
O=I
BOU I=Aovua
L**L=MJLVd
slawal= W) Iia
***s.~l1IA idhI****=AOvff
*****ILnVd~ldH****=81J.~f
~ 1LfidMi*****=sictaj;
**.**fl'VA JdidII.****=H
96*0=(S) A
titaiO= (t)a
8t o=(Z) I
82180~d 10 CalIIOd~g SV SZO'IVA aL
VSd (S'raVZS
849aMddgE .1VZV
sassva RRORIMI
(Oooo) J(00O~ihoaijl(ooooa)jLsL O(oor)xii~l loisiauio
(OSO)VSd ikoismavia
(00Ja'(OOZ*oi)Vea'IV*(OOZ)R KOISHRUIG
anivYadNRaSHVHZ3SIa IVfVDZ=L
I'IZZON I V AV~dS JO aHOIVHadUlI=*Z D
smoILvigLI ao Ra~0waK=aIflI 3
m~l agnaadwaJ. IV ldIveJlgz=LH 3
HaRM SSVd=l a
oijD1Iu kaaoo=a 0
a3JdS UMIEOS(12dSft
zanivaazwai aing zaAagi 3
1=1+ I
C BEGING PASS CALCULATION WITH UPWIND NODULE
10 J=J+l
IF(IEQ.1) TWBL(J#.r)=F(J) (Tfil 1 (1) -TWB) +TWB
IF(I.GT.1) TVBL(J,,I)=P(J)*.(TRIX(1-1)-TVB)+TVB
NTU=O. 16+ (0.053) *WSPEED
IF(r.EQ.1) TW=TRIX(l)
IF(I.GT.1)
TS--TN-30.0
ITVBL=TVBL (JI,I)
ITVBLI=ITWBL*l
PSATBI=PSA(ITWBL)
PSATB2=PSA (IT'VBLI)
ITN=Tg
ITNI=ITN+l
PSATNI=PSR(TTN)
PSATV2=PSA(lT*l)
TITN=ITN
TITVBL=ITVBL
PSATN=PSATNI+ (TV-TITV)'* (P5k-T52-PSATV.l),,-.;.
PSATVB=PSATS14-(TlBt!IJ't':t), ;IITVB'L 'f.P AT824' KTBI)
HTN=0,935*(O.o24*TW+(-ow622*PSATN/-(14*7-PS ATN))*0061,8+0e44*TY))
HTVBL=0.935*(0.24*?VB.L(Jol )+(Oo622*PSATVB/(14.7-PSkTWB))-
I *(1061,8+0.44*TVBL(Jrl.)))
ITR&TZ=0
C BEGIN ITERATICN FOR HEAT TRANSIFE18. CCALCUL:AT1611'
30 TS&VE=TS
ITRATZ=ITRATB+ I
ITS=TS
ITSI=ITS+l
PSATS I =PSA (ITS)
PSATS2=PSA (ITS 1)
TITS=ITS
PSATS=PSATS1+(TS-TITS)*(PSkTS2',p 1. '1
HTS=0.935*(0.24*TS+(O 0622*PSATS/(14..7-- PSi-'At.t))..*(1061.8+0,44*TS))
TS=TN-NTU*((HTS+HTN)/,2.0-HTUBL),
log IIU3 ost
8 OJl 05
osI. ox o!) (sassvdeoabi)ai
anuiJmioD 001
((Awi1 )d'vw a(I 'I)J.ta-M1x/I i ) W I I I
((I~n~-1*soI'll)V 8 fl*8i*10If l-ItLoa-ot ) (I) lt
aouvt='iw 001 Oct
oo = (I) lKimJ
(E'SLJ* SI 21CZOJVdURJ. GRINl LKSRHd an~ HtE)Lvlaaoi ZL
L 1 xI (oa 0 Sol ux) Ji
ooo=(tOa (os~a;a)ai
(i'x) VHd1lV+VHd IV1--V HdT IJ
aoumst= i OL oa
0 0=vHdTvI
102I1R03 09
SS~d H3U U0i SISSO'I RAUV8oodvAa aaROS 3
09 oa. os (moax*0ir)aI
(E*Sti SI WElDI I
S.S~d 011 IJOL"EI 808 IV Ra103 cIo VddS alll H8l)LYII)Is
OE 01 09
os 03; of) (tioovuazxo)&
(s-L-ZAVS.) sGv=ULLJIUa
TCC=TMII(I)
160 CONTINUE
WRITE(6,175) TCC
175 FORMAT(36H CANAL DISCHARGE TEMPERATURE EQUALS ,F15.3)
STOP
END
C THIS PROGRAM PREDICTS THE THERMAL PERFORMANCE OF EVAPORATIVE
C MECHANICAL DRAFT COOLING TOuERS-- ECDAFT-
C KAL=KA/L
C JKEEP=RETAINS VALUE OF J
C TWI= TOWER INLET TNMPERATURE
C ALPHA=PACKING EMPIRICAL COEFFICIENT
C BETA=EMPIRICAL PACKING COEFFICIENT
C AIRG=AIR LOADING LB/HR-FT**2
C ACELLW=NORALIZED CALCULATIONAL CELL WATER LOADING AREA
C ACELLA=NORNALIZED CALCULATIONAL CELL AIR LOADING AREA
C RLG ==RATIO OF WATER FLOW TO AIR FLOW IN EACH CELL
C NOITT=NUNBER OF ITERATIONS
C N=SQUARE ROOT OF NUMBER OF CELLS
C HEIGHT=HEIGHT OF PACKING
C GNTU=KA/L
C FNTU=KAV/AIRG
C DNTU=NTU PER CELL
C TO=OUTLET WATER TEMPERATURE
C PS=SATURATED VAPOR PRESSURE AT TOWER WATER INLET TEMPERATURE
C H=ENTHALPY OF NIOST AIR AT WATER TEMPERATURE
C TS=SATURATED VAPOR PRESSURE AT SPECIFIED METR. CONDITION
C HA=ENTHALPY OF moIST AIR AT SPECIFIED METR.CONDITION
C J=NUMBER OF ROW ACROSS, I=TOP ROW
C I=NUEBER OF ROW ACROSS, I=AIR INLET SIDE
C TI=WATER TEMPERATURE
C HW=MOIST AIR ENTHALPY
C KC=ITERATION NUMBER CHECK
C TV2=TEMPERATURE OF WATER AT THE EXIT OF AN INCRSENT(CEL)
C TWBAL=TEMPERATURE OF AIR AT TOWER EXIT
C DH=ENTHALPY DIFFERENCE OF THE AIR BETWEEN THE INLET AND OUTLET OF A CELL

A, I
C DH1=ENTHALPY DIFFERENCE BETWEEN WATER AND AIR ENTERING A CELL
C HA2=ENTHALPY OF HOIST AIR AT THE EXIT OF A CELL
C THIS PROGRAM HAS INCLUDED THE VARIOUS PARAMETER VALUES FOR THE
C FOLLOWING TOWER
C FILL WIDTH=36 PEET (INCLUDES BOTH SIDES)
C FILL HEIGHT=60 FEET
C UNIT FILL LENTH=32 FEET
C PAN DIAMETEB=28 FEET
SUBROUTINE TOWER(J)
DIMENSION TWBIX (380) ,CTWOUT (380)
DIMENSION HW(30)v, TW(30), PSA(150)
COMMON/TOWTEM/TTOWIN
COMBO N/VET/TWBXX
COmMON/EFF/CTWOUT
CONON/ATMOS/PSA
COMMON/WATER/WATERLWI
REAL KAL
JKEEP=J
TWB=TVBXX (J)
TWI=TTOWIN
53 CONTINUE
ALPHA=0.065- (TWI- 110. 0)*(0.000325)
IF(TWI.LE.90.0) ALPHA=0.0715
BETA=.6
AIRG=1692.0
KAL=ALPHA*(WATERL/AIRG)**(-BETA)
HEIGHT=60.0
GNTU=K AL* HEIGHT
ACELLW=32.0
ACELLA=53. 3
ROCELA=ACELLV/ACELLA
RLG=ROCELA*VATERL/AXRG
NOITT=5
CONST=7.48 1/60.0/62.0*10.0**9
CONST1=0. 124683/62.0
PATM=14.7

It
WIR ARCI - (I) 46 = ) l0
Iouvoaxi ULiSMva ivae ama
~~xoK 01 09
I" 8hL 5 Lila 31) 11
0 *Z/ (ea,*uuu) =&ieo
flJIN(10 s Z (Ha- H-ZImn. I UCI) =ea
(z1J;*tb 0.8 * 190) *(Sd -IIJ.d) /sd*Z 9 Oezl3*bU 0=ZM
m Rz/ea - (I) BL=z aI
e~IDNI*V i/I ea=ua
svozzviMaiI ss 01 -oaia
sgi=L! tiOL oci
R= (I) ail 00i
(u1.*11ttiO48090 0* (SI-Vl1.d) /S1.suZ9oea11.*tZ o=VH
W11111i) W(.) VSd- (1 +-11) 151) U1.) Vgd=S1
(11Im1) Wi) VSd-A (i 1) VSd) + (J61) VSd=Sd
R=II =.
-4I
C1011
OAJ6 (daxc 211MZ fOouJL
Ui) aLO.=R £01ol&-nz o
31*L:I £01 0a
0 OL 0=0
ZVH=ZZVH
(Zg~~tt +01got * (s-lu,) /S*Z 0 **l*t * O9J=Z5VB
(rs~l-reaJ * (rsLI) isa-t (I smiAL) vSd) + (RsAI) !Sd=Sd
nSL=2IUMJ. 0$
0IL~SA~L 0
oz oilo0
0 S+Z9lBL=?l
(zallaau)a ds). (rULI)V Sd-Sd
ZURI~=ZGftI oz
S5083V BOB eDVR 8OA 9filvi~adval fl'W Lag xSflvHxa HUv RlusalG D
9841+H=&i L0t
(I- (I) Ila) *(JI) 'VSd- (1.I) i Sd) + (JIi) VSd=Sd