Professional Documents
Culture Documents
o
39!
CRANFIELD INSTITUTE OF TECHNOLOGY
1
Ph. D. THESIS
Session 1988-89
SOLAR REFRIGERATION
Evaluation Of Technical Options and
Design Of A Solar-Generator-Adsorber For
A Novel Adsorption Refrigerator
(Volume
SEPTEMBER 1989
ABSTRACT
have been developed and used to identify factors which influence its
areas where insolation is high and sunshine hours are long. It has
the day, may be more economical and eliminate the need for a battery
storage.
coordinated effort, biogas plants. Biogas can then fuel the modified
carbon and the metal container. The idea was practically implemented
The seal between the metal container and glass imposed practical
difficulties in that design but the tests proved that the idea had
adsorption refrigerators.
iii
ACKNOWLEDGMENTS
I would also like to thank all SME staff and other friends who were
My thanks go as well for the test area staff who afforded me much
This work could not have even started without the enormous support
go to my wife who, despite her ill health, was able to support me,
able to sponsor me, for over two years, to carry out the research
SUMMARYOF CONTENTS
VOLUME I
1 Introduction 1-1
VOLUME II
APPENDICES A-1
J -7
V
ABSTRACT i
ACKNOWLEDGEMENTS iii
SUMMARYOF CONTENTS iv
TABLE OF CONTENTS VOLUME I v
LIST OF FIGURES VOLUME I viii
LIST OF TABLES VOLUME I xiii
5.1 5-4
A area m2
AC collector area m2
AMx air mass x
Cx specific heat capacity of material x Jkg'1K'1
d diameter m
x emittance of material/surface x
Gr Grashof number
g gravitational constant mS'2
he convective heat transfer coefficient Wm-2K-1
hr radiative heat transfer coefficient Wm'2K'1
I average monthly irradiation Jm'2
kx thermal conductivity of material x Wm-1K-1
L length m
m concentration of refrigerant kgkg'1
efficiency
6
slope degrees
wavelength micron
Al kgm-1S-1
absolute viscosity
V
kinematic viscosity m2S-1
density of material x kgm-3
SUBSCRIPTS
a ambient, air
b battery
c charcoal, condenser, collector
e evaporator
eff effective
g generator
i inverter
m methanol, maximum
sky sky
t top
CHAPTER ONE
Introduction
1-2
1.1 Objectives
the EPI are at the last level of the cold chain i. e a rural health
prices of oil makes these fuels more expensive and beyond affordable
out in [4], contaminated oil can block the supply pipe, or a break
in the pipe can be a potential fire hazard, the wicks need constant
the market for quite some time. But high initial cost, recurrent
and the need for high technology photovoltaic cells and associated
refrigeration
The report has been divided into nine chapters; each one fully
each one.
Chapters two and three present the review of refrigeration and solar
prototypes and commercial units has been presented at the end of the
chapter.
Chapter six deals with the further evaluation of the three selected
options. It is subdivided into three parts. Part one deals with the
has been developed. The model enables one to work out the size of
weather data.
assessment of the results has been carried out. The results have
been compared with those from other sources. Arguments have been put
forward when the results differed. The anomalies arisen out of the
adsorber (SGA) has been tested and the results reported here. The
The design of the SGA was not satisfactory and alternatives were
proposed.
whole work described in this report and takes inferences out of it.
Proposals for future work has been put forward in this chapter which
REFERENCES
CHAPTER TWO
Refrigeration : An Overview
2-2
surroundings.
23 REJECTED
TO
SURROUNDINGS
31 CONDENSER J2
W34(+ve) (-ve)
/W12
EXPANDER COMPRESSOR
41
HEAT LEAKAGE IN
EVAPORATOR
041
THROUGHINSULATION COLDSPACE
(a)
TEMPERATURE
T2
w
ºCETEMPERATURE
w
Ti
a.
W
I-
ENTROPY
(b)
Fig 2.1 Carnot refrigerator: a) system layout
b) thermodynamic cycle represented on a T-s diagram
2-4
that the COP reduces greatly with the increase in heat rejection
refrigerator.
practical systems.
2-5
28
26
24
22
20
z
18
0
16
II
0
14
zw
12
w
0 10
2
20 30 40 50 60 70 80 90
T2 (°C)
HEAT REJECTIONTEMPERATURE,
expansion devices and the rates of heat transfer which are expected
wicks onto the side walls and equipped with a water storage tank had
Being an open cycle process this needs a cheap and readily available
and b) in an expander,
All these changes have been forced upon by the practical working
23 REJECTED
TO
SURROUNDINGS
3 CONDENSER
2 /W12 (-ve)
THROTTLING
VALVE COMPRESSOR
4 1
HEATLEAKAGE IN
EVAPORATOR
THROUGH
INSULATIO (241
COLDSPACE
(a)
W
cx
cx
W
W
I-
ENTROPY
(b)
Fig 2.3 Vapour compression refrigerator: a) basic components
b) thermodynamic cycle represented on a T-s diagram
2-9
the plant. Thus, the expansion process 3-4 is not isentropic but is
The reason for moving state point 1 onto the saturation line is
state 1 in the wet region as shown in fig 2.1b. Secondly any liquid
refrigerant, which will be part of the wet vapour, passing into the
compressor.
modified (see fig 2.4a) for practical reasons. The factors which
2. ub.
2-11
Cf.
W
lLJ
I--
ENTROPY
(a)
SUBCOOLING
PRESSUREDROP
3' 2'
W
V)
W POLYTROPIC
0. 4' COMPRESSION
PRESSURE
DROP
SUPERHEATING
ENTHALPY
(b)
Fig 2.4 Actual vapour compression cycle represented on
a) T-s diagram b)p-h diagram
2-12
The p-h diagram of fig 2.14b clearly indicates that compression work
1'-2' is greater than the work done in process 1-2. This lowers the
region of 3-5.
Certain liquids and gases can be dissolved into certain liquids and
refrigerator.
heating the solution and then throttled and evaporated in the usual
shown in fig 2.5a. Comparison of fig 2.3a and fig 2.5a reveals that
pressure gradients between the two sides are balanced through two
EXTERNAL
HEATO9
GENERATOR
ao
00
3 CONDENSER
THROTTLE P
VVALVE
WUOP
ÖRBER\I
4
HEATLEAKAGE
THROUGH "xx
INSULATION VNýwv
De QS
THIS PART REPLACESTHE
COMPRESSOR OF THE
CONVENTIONALREFRIGERATO
(a)
W
F-
ENTROPY
(b)
Fig 2.5 Continuous vapour sorption refrigerator: a) major parts
b) thermodynamic cycle represented on a T-s diagram
2-15
then fed into a condenser to complete the main cycle, while the
Tg
Og
HEAT
ENGINE
W9
(+ve)
TQ
We
(-ve
REFRIGERATOR
Oe
Te
The work produced by the engine is made equal to the work absorbed
unity.
The most widely used absorption pairs are ammonia-water and lithium
be produced using this pair. The one major drawback of using the
On the other hand the presence of any water vapours in the condenser
temperature.
RECTIFIER
CONDENSER öl VAPOURS
ANALYZER
GENERATOR
w
w E-
u ---J
w -O
--- If,
LO
Z
Y O cr-
N
LLJ
3 J
EXPANSION
VALVE z
a
1 ABSORBER
EVAPORATOR
PUMP
we find that the hot weak solution from the generator needs to be
cooled down before entering into the absorber. At the same time the
transfer the rejected heat from weak solution to pre-heat the rich
in fig 2.7.
This principle was used by von Platen and Munters [1], in what is
WATERSEPARATOR
b
FREEZER
EVAPORATOR
PRE-COOLER
BUBBLE
PUMP
GASTGAS HEAT TUBE
.
EXCHANGER
CABI T
EVAPORATOR
ABSORBE
HEATING
PIPE
ABSORBER
VESSEL
STRONG
saunON
Electrolux for their product are + 30 on the tilt and + 10% on the
This cold mixture of hydrogen and ammonia flows down to the absorber
where ammonia is absorbed into the weak solution coming from the
generator and hydrogen rises up towards the evaporator. The two gas
streams, the cold mixture of hydrogen and ammonia and the hot
four processes are carried out in two phases. Figure 2.9 illustrates
the principle.
In fig 2.9a vessel 'A' contains the sorbent and vapour 'mixture'
are driven off and are condensed into the vessel 'B' after loosing
latent heat to the ambient. This is the first phase of the complete
The fact that generation and sorption processes are taking place at
different times can lead to big advantages. This means that there is
2-24
VAPOURS
PROCESS
GENERATION
(a)
VAPOURS
COOL
PROCESS
REFRIGERATION
(b)
Fig 2.9 Working principle of intermittent sorption refrigerator
a) generation process b) refrigeration process
2-25
following advantages:
vessel, and
emf was produced in the circuit when the two junctions were
gradient along it and the current flowing through it. This effect
At the turn of the century Altenkirch [3] gave the first serious
particular application.
but with the decrease in size the cop of the latter falls.
rni n irr
(a)
).C. CURRENT
M HEAT ®COOLING
SINK SURFACE
(b)
Performance Characteristics
Thermoelectric Module 801-2003
Cambion Inc. Cambridge, Massachusetts, USA
24
22 .I I
IQ a
v
W
ýý 30
U
lac 1.ý
Cd
25
H
ö 1ý'
20 20
1 30,
U' I5
r-ý
7c 0
'or 40 0
0
0 ý. ý
1 I0
50" -4
45
30 5
/I X. "w
ýK 60i,
n,
e5678 9
CURRENT (amps)
parts and thus a longer life this has been applied in the military
2- the hot and cold junctions swop their role if the polarity of
when the inside temperature goes below (or above) the set
2-31
a mean.
is about 0.01 bar and the corresponding specific volume is about 129
m3/kg. Such high volume flow rates are not suitable for reciprocating
compressors. This has led to the use of steam jet ejector in place of
pressure, into the converging part of ejector pump. The high speed
steam entrains the water vapours from the vacuum chamber. The
LOWPRESSURE
VAPOURS
1ý1
3456
2 INTERMEDIATE
URESTEAM
MIXINGZONE
DIFFUSER
HIGHPRESSURE
MOTIVESTEAM
cal
EXPANSION
OF
MOTIVESTEAM
N
In
w
0..
1345
POSITION
ALONGTHEEJECTOR-PUMP
(b)
in a pressure-distance plane.
shown in fig 2.13. High pressure steam, typically 70-700 kPag [6], is
entrained and mixed with the steam from the boiler. The mixed steam,
then diffused through the divergent part of the venturi until the
These refrigerator units require only the heat input to operate and a
BOILER :
-öo-ö
/HEAT
INPUT
FEED
EJECTOR PUMP =
PUMP
-O-
-o -p r
EVAPORATOR BLEED
VALVE /CONDENSER
HEATGAINED HEATREJECTED
FROMCABINET TO ATMOSPHERE
BOILING
LIQUID VAPOUR
ö----
uauio 7]
-_c =o
c F.,
REFERENCES
London, 1964.
1972.
CHAPTER THREE
an observer on the earth, about its axis every four weeks. However,
it does not rotate as a solid body; the equator takes about 27 days
and polar regions take about 30 days for each rotation. It radiates
The amount of this energy reaching the earth is defined by the solar
constant Go. It is the amount of energy from the sun, per unit time,
solar constant is 1367 Wm-2 [1]. Recent satellite and rocket data
[4,5] has confirmed that the previously used 1353 Wm-2 [2,3,77 value
instrument calibration errors [6]. The data confirms that there are
daily and monthly variations of not more than +0.25% and changes
systems.
3 -3
TABLE 3.1
constant has wavelengths shorter than 0.3 micron (i. e UV and X-rays)
and the rest 54% is in the infrared (IR) region. 96% of the solar
when the sun is directly overhead and, at the sea level, is defined
changes as the earth rotates diurnally. When the angle between the
zenith and line of sight is 60° the distance traveled by the beam is
[7)) shows the solar spectrum under 'air mass 1' condition. Parts
2.4-
2.0-
Ei
E 1.6
1.2-
E
X30.8
~
w 0.4
v0
0.2 0.6 1.0 1.4 1.8 2.2 2.6
WAVELENGTH(µ m)
(a)
E
=- 2.0
N
E
03
1.5 H2O
7O2IH2O
w
Li Absorption due to
atmospheric constituents
H2O as indicated
co
1.0 H2O
J
H2O
cr
ü 0.5 H2O C02
w
a.
H0 CO
variations due to earth's rotation both about its axis and around
the subject of the next sections of the chapter where some chosen
the plate) over which the solar insolation falls and is absorbed.
absorbing plate is higher than the ambient there are heat losses
from the front and rear of the plate. Thermal insulation is usually
placed behind the plate to reduce the heat losses from the rear.
Heat losses from the front are reduced by covering it with some
made from thin sheet metal or thin planks of wood e. g. plywood. The
c.
ýi
0
0
a)
a)
a)
41
0.
41
c0
-4
w
1-4
co
u
a
4
',0
m
a>
4)
c0
a)
w
4)
4)
U)
0
cý
v_ N
M
J0
to
H
w
3-9
channels. This ensures a steady balanced flow from the channels into
the header. The heat-transfer fluid enters and leaves through the
is the most used material for outer covers because of its superior
resistance to the environment. The glass sheets used are usually low
iron content and about 3-6 mm thick. The surface of glazing can be
a second cover under the glass cover. This protects the plastic from
and the coupled boundary conditions between the water and the tube
h- (za)eff - U1(Ti-Ta)/I
c
0
4.4
43
co
L
0
w
-H
ao
31
C)
4)
43
0
F-` x
4.
0
ýi a
0
4.3
cd
C)
a
t0
a
cd
M
M
00
This is because the convective and radiative losses from the plate
(the data for this figure has been adapted from [17]; the
N
O
a)
L
4)
N0
Ö
V
Cl
O L]
.
ÖQ
öE
C
°- - +I
"ö
ö[
6
d
O
a
DB
7~
8 °. S$SN30
.
(L) A3N3101443 1101031100
3-1ü
performance.
collector.
earlier. The absorber surface in this case can be a long and narrow
which heat losses from the absorber are equal to the heat gains).
Different methods for removal of heat from the absorber are employed
in different designs.
ID
N
6
V!
a m
L.
0
N U
N N
0 '-1
0
U
0
N
N
O
co
G].
ID CO
ö . -4
a-.
0
ö V)
a)
a
"E
c
a,
a,
w
öý
v
i
o_ w
ö
U
S
U
C
+1
U
0
4r
4-.
IR
0 M
00
C.-
N
0
0
0 °n 0 9$0 No0
40
(%) kDN3I0IJ33 801031100
3-16
Glass to metal
seal
Y
Fluid
Selectivesurface
Selectivecoating Vacuum
Entire spaceevacuated
Return
tube Feedtube
.
a) CorningGlassCompany'sevacuatedtube
Outlet
A -r--Inlet
Selectivecoating Vacuum
b) OwensIllinois Company'sevacuatedtube
mm. inner most tube was the delivery pipe through which the
the middle tube. This tube had a selective coating on its outer
surface and is hermetically sealed with the outer cover. The space
heat-transfer fluid through the ETC module. The all-glass ETCs are
that the failure of one ETC makes the whole module inoperative. The
ETCs are high performance collectors but are not used for
class by the way the heat is removed from the plate. Instead of
pipe is used to transfer the collected heat from the plate and
of the heat pipe transfers the heat by direct contact to the fluid
is presented in appendix A.
3-19
(1) Condenser
(7) Getter
A -A
lain
material was used with cuprous oxide to develop a solar cell. The
silicon solar cell was reported in 1941 but their widespread use as
shell.
3-21
in fig 3.8.
band gap.
materials).
3-22
BAND
CONDUCTION CHECKVALVE
EXCITED ELECTF
Ec
STATE TRANS RT
hv, E
BANDGAP(Eg)
HOLE Ev
VALANCEBAND
with holes) and Fermi level lies within the conduction (or
valance) band.
The energy band diagrams for these cases are shown in fig 3.9.
when the energy of photons, making up the light, is greater than the
radiation.
3-2ü
E BAND
CONDUCTION Ec CONDUCTION
BAND
------ -Ed
Ef -- -ý.
Ef ---
Ev Ev
VALANCEBAND VALANCEBAND
(a) (b)
BAND
CONDUCTION
Ec
Ef
---- -- - -Ea
Ev
VALANCEBAND
(c)
Fig 3.9 Energy band diagrams for three principal types of semiconductors
(a) intrinsic (b) extrinsic (c) degenerate
3-25
acceptors are known as p-type. The most common solar cells are made
cells.
These are not necessarily separated physically within the cell. The
Fig 3.10 illustrates each type and one can see readily that cells
the larger band-gap material. Schottaky barrier can again have two
Table 3.2 gives the photon energy levels in different spectral bands
solar cell. This is one of the reasons for their lower efficiency.
3-27
O
+1
_i
rl Y
ab
sc
Y .+ º.
CE v
O7
N. ý M
M
4
Ri
C
Y
1. -
E 0 V)
V
fr N
.r
4 a)
U
ý0 4 C
0o u O L
N M cd
-u
s 0
ýp C
al
sM
u4
UE
aý -r+
19
r"N M
I V
k. 47 a)
L 0 A
C ýJý)
O G"C M0 0
- .r Y -4
-J U a
sC
ým C
0 s0)
1
CE 0
Oý w
V .y
4 I- u a)
r I!
ý S
ýý
IJ\/w ý' c C
a
c
V
a)
L
u
c0
4)
ILI C
iI U a)
Y CO M W
to 4)
0 0
ý./'ýJý L
RI
CL
ä; 44
ac GO
0
C m
a)
a a
V
r -1 Y
u
Y
"+d E"
0
u
41
V G C
O
bO
v w +I
I.. Y Lt.
II u
W C
x
.ýý,
r
0
x
a
c
3-28
TABLE 3.2
Different materials used in solar cells have different band gaps and
for band gaps and absorption coefficients was picked from the
diagram that slopes (on logrithmic scale) for some materials (i. e.
direct band gap materials) are much higher than others. This
between 50 and 100 micron thick whereas amorphous silicon cells can
O000N CO 0 In t- 0 N1 Lt
OOO l- Ln OOON
1
C
I-
OI
1 f/2 Lý O' M! LA O lý l. M OO
r- -
IO r- U1 N Oý M-N l- l- ýO O Oý O
1Z
O% N u1 C- 00 OO
6 OD
mIQ)
a
S0O -
Cý CO ýD Ln MN 10C)
Ft n1 '- ý- OOOOOOOOO ICe
IC6N
00 C1
. -4
Ü111111111111
. -i
rl 1? m 0% O% O% a% O% m a% ON HIvým
>61 a% CO l- %0 111 MN
I Ö
g1 ý- OOOOOOOOO 9.4+
1 1 z. oa
3-31
105
CuInSe Si
Ge
CdTe
Cu2S
a
r-
E 10,
v
u
CdS
W
Ü
U-
LL
W
O
u 103
GaAs
Gap
0
I
a
0
Q
102
10
1.0 1.5 2.0 2.5
PHOTONENERGY(eV )
literature, for some cell types, which are considered economical for
poly-crystalline Si 0.10-0.12
amorphous Si 0.06-0.08
references at the end of the chapter are provided for the readers
subject.
3-33
I-
w
u
w
w
ü
L
w
w
REFERENCES
6 Eddy, J. A., 'A new sun, the results from skylab', NASA
report SP-402,1979.
1985.
ASHRAE, 1967.
no 2, pp 59-72,1987.
3-36
14 Garg, H. P. and Datta, G., 'The top loss calculation for flat
1984.
5, pp 609,1984.
BIBLIOGRAPHY
Holland, 1983.
7 Open University, Solar Cell, unit TS251 (10 and 11), Open
at 18 months intervals.
19-21,1988.
4-1
CHAPTER FOUR
adsorption.
a) liquid/gas b) solid/gas
c) solid/liquid d) liquid/liquid
Kiselev [1], Everett [2], and Barrer 131 have classified the process
a solid. The solid in this case is termed as the adsorbent, and the
can take place simultaneously. The general term for the latter
(ii) chemisorption;
other.
independent of temperature.
4-5
is, however, small and thus the London's dispersion effect is the
major contributor towards van der Waals forces between the molecules
that its mean dipole is zero. However, such an atom may have a
another their total energy decreases and they are attracted to each
discussions by Ruthven [14], Gregg and Sing (15], Cookson [16] and
Bikerman [17] and the extensive references cited within these works.
4 -6
Lf(p)7T 4.1
4+-7
II type isotherms.
m= [g(T)l 4.2
The data used in constructing the fig 4.2 is from [18] and represent
isostere is
P= [h(T)]m 4+.3
cn
Ü Co
W
a0 0> CO 00 ý NN M ý
ý
W MMM-? aO ? ?
OC
0
. -1
O
92
tio d
"
C Co C"N
C)
. -1 G -. 0 4.3
m
ý1 L N L. O c0
0 cu ° CO m
U UX U -0
9. m " O
cu V- cyv v L.
p 0) x Z a) -4 tio (D -I x
. -1 i> NM 4.3 !C 00 0 0
to
CO CU. ^ cli O c
0C>+ to > O Cu U "-1
-1 U Co CO
UO +-1 .sU " +1 O +i
L 0
L +ý O 43 0 +1 0) +i U U a>
cd G1 UO . -1 UEL +ý L
X M ON 92 Co L Z cu
W -f
0 N Ox vI +-f 4-. Cu
U +1
O 0
COOON Co . -1 O tu Co C C U 03
O. i tO(D >"-4 OC 0
Ö to A 0 C
OOO
-
0 G) 0 (4.
1O
CC
93 020
O (d a) CC CN C 0
+4 ) L
C .-4 cd CO L. 0 C) '-40)0 +i r- C
0 >. 0.0 G) aN0 E+i N U
E 4) 41 +)
-4+f
L. -f ECL. 010 C +)
0
.0 > 02 O mr -4 3
912
4)
4wx23 0cm Ci 4 ca
1 U
'4
'i c ä c> ý V1
aý c Q) 0) Co .-1i
N .E-4 to -4 O . -4 0C >+ U .CO
4+
v,
-1 C T1 C L. 0 rl O +I O 4) C 4) "O
+-1 CU CO co
0 V! Co m +i 0) U4 91 41 d p4
N O L. o7 LdO >s +1 U
m 4 41 G1 a AC in
A m 4.3 C) CO L <O . -4 L d0 .i 41 d 4a m 1Q +-1 41 44
3O 4) C) +1
L .N U) , -1 CO 4) OO +-1 " L. "-1 o]
O L. O Co U>Z. E U) . 430'I V 4.0 09 c0 +i U
CIL
. C -1 W L. N L
a a) F 0 to V c in. CO.?
° 4.3 WC v] C) 0 c0 O L. .Cu OV al LC Co a u)
Co "-1 3 07 >º 4) EO 42 a CI >W
al C) vO C .CUOC
U) NUL. 42 O 'O ca
(0 L. . -1 CO L E 91
L L. C 74 OvLO O. O c0 O U) aý C o9 >
F1 .G CI
E-' O to OY G +) (O L >º U c0 U +-1 L 4) W U) N c0 OUH
+ý . -I E aULO CO C 0-4 +ý LL U) LWU U) C.
a Ua >s L OZOaL U) O CO
w ý, OU U) .1 7 4ý . -i «1
CLUa
aUU . -1 N LCO +ý O CO N U 4)
a 0
° e o 4.3 °' Co 0 :0 "' O mH
c ai a3 0 6- > 1 .C to
92 eo
-4 co
ä oý0Z0 0 (o0 42 Co H-40
a6 0) 4 q m- OE +3 C) La o"a "i a"
02 L7 .4 E7 U) c0 .CO Co CHL +3 .N
tu " +'1 0ý d0
.ý dO OON+. 3 O L. > Co HONL. Na 4) W
O> 4) W L. U L. N R) 0 44 O H 93 743 CO CN Ls.
L. c0 .4 +-1 0 +i . ' 4) 01 O 4) t. > «vE 4) Co UN +1
0 43 >s a on 4> 0 +i 0 Ho Cu O top .C v1
O
E- aOLWE.07 L. U O .4 CE 02 LOL
LCOa to a) .C CI C .4 U) 7 c0 4) N0 4) H0.000
V U U) Co c > 43 Z. CO v O c0 Q. c0 xN0 U) L.
+1 . -1
to OC Co +1 OCC C c0 C' U >s HC c0 aOOO
E+1 CO U. ] .C3 aa). 41 H> G> .C H H4) E+) C04) da
w
G 1A Uý
'' N N N
CL CL
V) CL Q. 0.
WW r' WW WW lü W r' WW
W H
ý+ H H >
H H H H >
E-+
4-9
175
933k Pa
150
53.3kPä
U.i 125
or-
0
ö 13.3kPa
U- 100
0
75 5.4kPa
W
a.
C3
ö so
0
O
25
J
O
slowly then rapidly with temperature. Thus the curves are concave
than superficial. Like the vapour pressure curves each point on the
different phases and allows one to calculate the amount the heat
d(lnp)/d(1/T) = -q/R
MM
EE m
E
U Li u
o L^ 0
c3
E
.. r
50
w
25
W
3.0
V)
w
cr
a
U- 2.0
0
r--
= 1.0
I--
Cf-
l7
On
.JVL
2.2 2.6 3.0 3.4 3.8 4.2
RECIPROCAL
ABSOLUTETEMP (ox 103)
Fig 4.3 Isosteric representation of adsorption data of fig 4.2
(a) linear scale for temperature and pressure
(b) logrithmic pressure and inverse temperature scales
4-12
(i) Homogeneous
adsorption sites.
(ii) Heterogeneous
(iii) Homotattic
Thus this section will cover the theories and models describing the
surfaces.
difficult. Many models have been put forward so far. Some are very
Thus the models discussed in the next few sections are equations
During second world war Langmuir put forward his kinetic theory of
range. But all of these equations either use the Langmuir equation
occupied or vacant.
4-15
v= vm[bp/(1+bp)] 4. b
dynamical principles.
4-16
This theory was developed by Polanyi at the same time when Langmuir
put forward his kinetic theory. The basic idea which forms the basis
by the equation
E= RT ln(pg/p) 4.5
where R is the gas constant for the vapour, T is the temperature and
temperature:
4-17
VAPOURS
SURFACE
vat?
ADSORPTION
SPACE
/
. ... i
ADSORBENT
[ac/aT]W =0 4.6
I
4-19
bo
V
,a
0
a)
O
M
ü
10
O
w Y+
*0
x:
Jy
CD ä
y
W
Co w
CY- 0
Q C,
w
Q
U
.. a
N
U
11
U
co
OD
. -1
Gc.
T-
(P1) 1VI1N310dNOI1d80SOV
4-20
above) .
b) the adsorbed layers beyond the first one are not affected
second and layers beyond is constant and equal to the latent heat of
the form
v Cx
u. 7
viii (1-x)(1-x+cx)
adsorption is used for the first layer [23]. The constant 'C' in
C= exp[(gst-hfg)/RT] 4.8
where qst is the heat of adsorption for the first molecular layer
X1 C_1
4.9
______ _ ___ + ----- x
V(i-X) VmC VmC
they assumed that maximum number of n layers are formed on the wall
before they meet the layers on the opposite wall. The derived
expression is
V Cx[1-xn(n+1)+nxn+1]
-- = --------------------- 4.10
Vm (1-x)[1+x(C-1)-Cxn+1]
Despite the fact that equation 4.10 gives a better fit to the
which is not true for many adsorbents (e. g. activated carbon). The
taken as benzene.
[30] as
empirical equation;
u-24
W= Wo exp[-B/ß2(Tln(p3/p))2l 4.11
in a linearized form
should get a straight line. The slope of this line represents the
In case of type A deviation (see fig k. 6a) the D-R curve shows two
effect but Rand [31] contests their point on the basis that their
data did not cover the point where the change of slope occures for
A deviations.
Type B deviation (see fig4.6a) is such that the plot is curved over
[33], for example, show this type of deviation. Marsh and Rand [33]
These deviations have been observed for nitrogen and argon adsorbed
3.0
2.6
2.2
0
r-
ö 1.8
1.4
1.0
E2x 106
2.
2.
2. ý
0
U3
0
J
2
..
2.
1.
20 40 60
E2 x 106
When n=2, equation 4.11 and 4.13 are identical and the quantities
D-R equation and many of the deviations pointed out above were
Polanyi's postulates are not verified and thus the D-A equation
P
ß=4.16
Pref
4-29
y= surface tension
nP
ß== where M= molecular weight 4.17
11ref
P= density of liquid
P
ref
Polar adsorbate molecules possess a permanent dipole moment and as
µ2
is =24.18
µref
affinity coefficient for polar and nonpolar adsorbates alike and the
polar nature. They also concluded that equation 4.18 or any other
[29,45,46, Z17].
4.5.5 Closure
one vapour from the experimental data of another for the same
adsorbent;
4.5.5.1 Implications
TABLE 4.1
Benzene 1.00 29
Propane 0.78,0.715 29,45
cycle. For clarity refrigerant saturation line and low and high
Clausius-Clapeyron equation.
temperatures set the pressure limits within which the cycle will
low and high concentration isosteres are located in the diagram the
The state 1 in the cycle is fixed by the evaporation isobar and the
cycle. Thus the isostere passing through this point is the high
- -J`frl
a
Y
r
W
ýý.
Lx
N
-- .
Lx
W
0
LL
O
-- --4
J
V
- -J. " O
Lx
ü
w
Lx
W
i výj
U7
' dye I W
Iý
aI
I
I
----------- _-
3änSS38ddO 901
4-35
state 3 locates the low concentration isostere. The two other state
evaporation starts.
Now that the cycle limits are completely defined, we can proceed
Q (Rln(pb/pa))/(1/Tb-1/Ta) 4.20
and
desorption, rises from T2 to Tg. Thus during this process the total
Q2-3 = ICA+CR((mh+ml)/2)](T3-T2)+(mh-ml)L((9st)h+(9st)1)/2l
4.25
....
Qs 2 Q1-2+Q2-3 4.26
Qc = (mh-ml)[hfg-CR(Tc-Te)] 4.27
Therefore,
absolute terms.
4-39
REFERENCES
pp 456-460,1959.
pp 99-124,1961.
1964.
London, p 59,1958.
1970.
1939.
p 151,1939.
54, p 705,1946.
P 133,1948.
York, pp 51-120,1966.
1976.
4-43
32 Toda, Y., Hatami, M., Toyoda, S., Yoshida, Y. and Honda, H.,
pp 565-571,1970.
pp 172-183,1970.
no 2, pp 195-203,1978.
p 69,1970.
75, no 1, pp 34-42,1980.
1928
USA, pp 479-485,1977.
PP 708-721,1927.
CHAPTER FIVE
5.1 Introduction
looks generally into the possible ways in which both these processes
are:
a) Manufacturing
b) Maintenance
country.
5-4
1
TABLE 5.1
c) Ruggedness
d) Service
Ideally speaking, once installed the units should not demand any
life (e. g. fifteen years) without any rigorous service except the
not exceed the range of 0°C to 8°C when fully packed with
when the ice packs containing water at 22°C are placed in the
be met.
5-6
The first thing which comes to light is that the required vaccine
the freezer which will be used for freezing the icepacks. Because of
drugs as well.
the basic maintenance and repair skills are available in this field,
which can be followed to convert the heat of the sun into ice. The
costs about $8.0 per peak watt (at 1984 prices) (3]. High capital
SOLAR ENERGY
This option has a big attraction that the solar operated unit can be
would help reducing the cost of the new system. A major disadvantage
are the ultimate target of this application. But photocells are very
rugged piece of equipment and these are not likely to fail during
cr-
tr
Oj w
cc N
ý Z
w
0
O
W u
DC L.
O 0
,. 3
N
N c.
W a)
ix m
cr. a c.
w
O a)
Fý- O
O u
0
in
a)
c.
a
cc 6
W 0
h- U
w G.
W
0
0.
CO
v
ca
4)
>- 0
Li cc 0
0
CL
OQ
>
OQ a.
0
O ix
R cr.
CC
0 0
I- a
e
0
cc
4)
AO
Gc.
5-11
input into 'cold' is only 0.1. The continuous operation of the unit
the sunshine hours only) is the heat leak into the cabinet through
The capital cost is ten times, and the storage battery requirement
times larger panel area and four times bigger battery storage. All
driven by the thermal energy derived from the sun. Among the
need either a heat storage to keep them running during no and low
balance the heat gains during the period of. inactivity (i. e. after
have the benefit that these would remain working during the longer
fig 5.3. There are three main components of the system, i. e. a solar
refrigerator.
organic Rankine engine seems to be the most suitable (i. e. R11, R22,
of heat engine and solar energy collector are opposed to each other.
references [7-10).
major factors which are not favourable to them are the lubrication
L.
0
.3
c.
CD
m
c,
a,
I.
U)
a)
C.
Cl.
S
0
u
0
a
6
L
a)
43
0
0
m
0
c
0
ca
a)
d7
al
L.
Ow
a)
c»
0
4)
e
d
U
03
-IC
M
LA
bO
"-1
w
5-15
values less than unity. CVAR can be driven thermally if the solution
Due to the absence of any moving parts, these refrigerators are very
range of 0.2-0.25 but these can perform well for their lifetime
ü
e_
d
T O
ý Cl
i LA
Ö
co
O
00
c.
w
O
[.
N :.
r
a
6
0
V
R.
O
a
ca
uC
" a)
ui
cr- D
~ä
c L.
CSC
W a) a
W ++
c-4
~=
Jö cc A
0 f- a, *v
"- =0C:
Oa
0 L.
W[
-ý
a . -4 a>
0-0
L. C
fa aa) ö
>U
0
00
CC
tii +-1
bO
O ca
C% +i XG
4) V
co
w +4
>S
t[1
00
4-1
w
0
ao
.tOT 'C -
C= Ö
96 1- V- OÖ
33NYW80d83dd0 1N3IJ1dd303
(11Y3AO)
5-17
and thus very reliable. The COP is generally below 0.6. The sun is a
temperature.
Table 5.2 enlists some 37 of all three types of those sorption pairs
TABLE 5.2
level. Thus this will be one of the deciding factors while choosing
sorbents would reveal that solid sorbents have lower specific heats
0.92 of zeolite with 4.18 for water). Further it was observed that
TABLE 5.3
Density at 25°C
Specific heat at 25°C
Latent heat at -100C
use metals which have low specific heat and low density. But there
which affects the total mass of the metal used is the system
(or vacuum) is perhaps more desirable. This will help reducing the
sorbent and therefore help reduce the overall size of the plant.
important. As the lower COP will demand a higher heat input to the
absorbents. The results from [15] and [16-18], shown in fig 5.5,
sorbents.
5-23
0-25
020
W
U
9
015
im
W
a
U-
C)
F-
W
ü
U- 0.10
w
0
Li
Q
J
O
(/)
0.05
0
80 90 100 110 120 130 140
(*C)
TEMPERATURE
COLLECTOR
Fig 5.5 Solar COP of IVSR system for various sorption pairs using
a double glazed selective surface flat plate collector
5-24
From the list of sorption pairs cited in table 5.2 activated carbon,
System' and the patentee was A. T. Bremser [52]. The designed plant
collector (1) and the rich ammonia mixture while passing through it
number 3, into the generator (4). The ammonia vapours from (4) were
fed into the condenser (5). The weak solution from (4) was partly
fed back to the solar collector and remaining went to the absorber
Platen-Munters' unit. No evidence has been found that the unit was
ever constructed.
a pipe which was placed at its focal line. Two weight driven alarm
clocks were used to keep the collector track the sun. The steam
proposed.
5-26
N
Lt
u
x
O t.
H
d
N
y
E
O a)
a L.
6
44
O
C)
a)
a)
a)
a)
ca
0
co
a
0
C
tkO
0,
a)
10
ýv
a)
co
0
a
0
L.
a
bO
5-27
large parabolic mirror. Very high cost of equipment and the low
shell. This was lined with aluminized mylar polyester film and
paraboloid was 45.7 cm. The cooling space of 0.064 m3 was maintained
overall solar COP was reported as 0.14 for aqua-ammonia and 0.08 for
5-28
been the low thermal conductivity of the solution. R12 also needed
higher than the absorber, and a pipe, C, connecting the two which
pipe; ammonia was driven off into the generator and weak ammonia
period.
the system Trombe and Foex were able to reduce heat absorbed, from
U: -- : __ E
JD
i
ict
iC
"=-
A ammonia reservoir
B liquid heat exchanger
C heating tube placed at the focus of
the cylindro-parabolic collector
D boiler
E condenser
F water tank
G cooling coil for the condensate
H ammonia collection vessel
I evaporator coil
J ice container
K cold chamber
flowing. A storage tank for hot oil was also connected to the solar
collector for night running of the unit. A flat plat collector was
also connected in the circuit which preheated the oil entering the
heated and maintained at 1800C by the circulating hot oil from the
separately (fig 5.8). The generator was combined into the solar
106.7 cm, painted black. There were three glass covers supported by
cork strips and the back was insulated with cork board. Six steel
collector sheet, made the generator. The absorber and condenser were
L.
0
co
c.
mrn
L, u
4-I
co L.
"1 O
C to
OA
E R1
LL
O
L+ý
O co
+) L
ov a)
3G
a)
4-. 00
O
L
CO
ON
4-) aD
cd -4
cO
4) O
to
CDb
L. G)
ac
a)
L, ü
UO
+l c)
cc 0
6
O bO
C
U +1
d1 >
co
h0
rx.
--J
"ýJ ;IE.
/J {
1"
N
5-32
in the absorber at a pressure of 3.0 bar. When the sun set the
generator cooled down gradually and the pressure dropped from 8.0
bar to 3.0 bar. At this point valves between generator and absorber
opened and the rich solution transfered to the generator for another
194 W per day with a cooling ratio of 0.58. The proposal did not
1
___-_.....
r-ý===__-_ yI
iI
Iý
II 0
I u
I
co
00
I "i
0
II
±
I
10
J
N
P:: 93
0
a
0
a
0
43
a)
m
1.4
c.
a
a)
L.
co
4.
0
m
ca
u
4)
ca
a>
U
N
ON
LA
to
[s.
5-34
Condenser, evaporator and absorber were all shell and tube type
maximum amount of ice during the day. The generation temperature was
24 °C, with an over-all COP of 0.1. The system worked on cloudy days
solution pumps.
5-35
319
uo N
a)
rl
tS1
L rl
f N 0
C
d L.
L 4.3
ýv co
a
CL
a)
0oN
Q d1
a)
0
CL 4.3
L
a)
d. bO
C)
L
4.4
0
c0
L
bO
cd
-4
V
C)
C
14
c.
L++ 0 ON L.
EL
v-u
V O
46 Lo "-o vO4
U;
"- ac .+ m
0= bO
N r- L
üc t.
3
5-36
header at the top (see fig 5.11). Thin copper sheet was soldered to
the tubes and the whole assembly was placed into a wooden box with a
was 0.07 and 0.26 respectively. The system with Sodium thiocyanate
-OFF VALVES
5 6
CONDENSER-
RECTAFYPJG EVAPORATOR
9
PRESSURE C ON-OF F
GAUGE--% VALVE
2
a
ING LINE UPPER NDENSATE10
HEADER DRAIN UNE
D
-OFF
VALVE
7
VAPOUR
COLLECTOR RETURN LINE
TUBES - DURING REFRIGERATION
4
EXTERNAL
RETURN LINE
LOWER HEADER
3
hot box. The unit was tested in summer 1970 and the results
that 2 kg of ice was produced. It was concluded from the study that
pair. The solar energy collector, which was also acting as the
The evaporator, whose area was kept large for natural convection
heat transfer, was placed into the ice generator. Two problems were
energy collector area was 0.5 m2 whereas the area needed to produce
sufficient gold for 24 hours was 2 m2. The paper is very brief and
tank was introduced in between the solar energy collectors and the
The condenser, the absorber and the rectifier were water cooled and
the heat from the cooling water circuit was removed through a
5 oC in there.
Three pumps were used in the system: one to circulate hot water in
water from the cooling tower and the third to pump the rich solution
The plant was tested in April and May 1977. It took 2 hours to cool
circuit used R-12. Water at 101°C from the solar collectors was used
for the Rankine engine and 40% for the solar collector, was 0.2.
sense that it did not require any auxilary power. Circulation of the
strong solution between the absorber and the generator had been
the generator. The week solution was fed to the absober by a bubble
5-41
That in turn called for the water cooling of both the components.
with a cycle COP of 0.4. Despite the very good performance the
areas where water is scarce. For the optimum operation of the bubble
of the bubble pump seriously and may even stop it functioning. This
make about 100 kg of ice per day. The reports and papers cited above
The latest plant, which had been tested and reported in [79], is
PLAN VIEW
Ammonia
Ferrocement Receiver
0I234 Tonk Ice box Containing
Containing Evaporator
Scale (metres ) Condenser
`Water
`\Walter Purging
Trap
`-, Vapour Line
Liquid Seal s. Purging Line
- Purging Line
Solar
Collector
Absorption Distributor
Line
-- _
r
SIDE VIEW
Firrocernent tank
Position for auxiliary mirrors.
Liquid contak*q condenser
Mal
Solar collectors -- -, R+cciwr
_I Purging
.
line
r -+
0I23 Water
purging
trap
Scale (metres )
two rows. The concentration of the strong solution was 46% and its
coil and the ice cans. The operation of the plant was intermittent;
ice but it could produce only 31 kg. The reason pointed out were
Despite this discrepancy this was a successful design which had room
kg of ice per square meter of collector area. The units are produced
in three sizes, i. e. 25,50, and 100 kg ice per day and their
A lot of work had been reported regarding the use of metal halides
the outer tube had a selective surface and absorbed solar energy.
The inner tube with 6 longitudinal fins helped to carry away the
condenser with the tubes and filling them with R-114. The air cooled
tä<
Mrs
ýý ý' ,ý, ý,, "2ýMii
ý
ý°, ý sý 'ý'zý
.
..... ^ ý'r
The COP of the refrigerating cycle was found to be 0.34 and the
0.28 [84]. The major drawback of the design is the inability control
The price varied according the type and capacity of the unit and was
in fig 5.16. There are three main parts of the the device:
tubes were filled with calcium chloride granules and ammonia and
above ambient.
5-48
CYLINDERO_PARABOLICCOLLECTOR
c
Fig 5.16 Calcium chloride ammonia absorption refrigerator
manufactured by Comesse Soudure S. A., France [87]
5-49
functions: a) the ice production; about 10kg per 24 hours with 4.8
storage.
-320C and -29°C. This indicate a need to improve the heat transfer
France.
zeolite (zeosarb 5.0 A) and made 6.8 kg of ice after a sunny day
the day and cooled an insulated box of 113 dm3. The refrigerator
worked on a cloudy day as well, though ice formation did not occur .
The refrigerator was not suitable for producing ice. The temperature
Activated carbon and methanol adsorption pair had been the focus of
the plant is shown in fig 5.17. The main parts of the ice maker
were:
(maxorb). Fins, to enhance the heat transfer from the front into the
ý, .
.,.
,i
4 COL.EGZ'f CF AREA62
of the box. The box was placed in an insulated casing covered with
single glazing.
2) Two air cooled condensers were part of the ice maker circuit each
communicating with two of the four solar collectors. These were made
had enough volume to contain all the methanol desorbed during the
day. The methanol inside the evaporator boiled and absorbed heat
from the water to be frozen. Every morning the ice was removed.
The results from the tests on the experimental ice maker were
showed that when the front of the collector reached its maximum of
107°C the adsorbent was at 92°C. This indicate the high value of
contact resistance between the front wall and the activated carbon
granules. The maximum ambient temperature on the test day was 23°C
close (3 mm apart).
percent and the maximum insolation on the day of the test was
between the wall and the adsorbent at that instance was 15°C which
pair had been manufactured by a french company BLM [99]. It was very
Fig 5.18 shows the essential parts and general layout of the plant.
Aa
y_
..
" ýýýý,
REFERENCES
no 1, pp 9-14,1982.
pp 57-89,1983.
pp 89-95,1988.
vol 6, pp 309-316,1982.
5-56
paper no 73-WA/So108,1973.
1982/3.
India, 1984.
no 4, pp 253-260,1986.
Australia, pp 500-506,1983.
1984.
1981.
Brussels, 1982.
1970.
231-236,1984.
5-61
1978.
'A heat pump and thermal storage system for solar heating and
pp 239-249,1985.
Vol 2, pp 749-760,1981.
5-62
1981.
1933.
pp 527-531,1983.
Italy, pp 788-794,1980.
5-63
PP 731-734,1985.
1933).
5-64
1957.
November 1958.
Paper no 73-WA/SOL-6,1973.
67-WA/Sol-4,1967.
Paper no 72-WA/Sol-3,1972.
pp 83-104,1975.
1975.
Thailand, 1984.
India, 1986.
Denmark, 1983.
PP 197-216,1984
pp 479-485,1978.
May, 1988.
5-70
3, pp 259-264,1986.
pp 1595-1606,1987.
PP 303-310,1987.
5-71
March 1989.
6-1
CHAPTER SIX
PART I
Photovoltaic refrigerators
6-2
desirable one. For this purpose the size of the plant and area of
The cooling capacity of the unit was calculated assuming the- cabinet
sphere, offers the least surface area per unit volume. The cabinet
which were:
(i) All the six walls of the cabinet had the same thickness
(ii) All the cabinet walls had the same heat transfer
characteristics
(iv) Ambient temperature was 32°C and 43°C (night & day)
6-3
Heat transfer to the cabinet takes place through all the three
conductivity k.
The heat transfer which takes place from the external to the
The radiation at the inner surface was neglected but at the outer
cabinet surface and the ambient was less than 50°C the radiative
h= dE 1T äv
r
The overall heat transfer across the cabinet walls was then
G-5
24
22
20
18
(/)
Z_
Q 16
14
w
12
Z 10
w
Z8
0
ý6
Z
w
4
0
0.02 0.04 0.06 0.08 0.10
INSULATIONTHICKNESS (m)
Fig 6.1 Heat gains through insulation for a 30 litre cubic box
6-6
would be 3.3 W.
= 0.2 W
inside air is replaced with the warm and humid outside air. To
quantify the load imposed by the warm air is again difficult because
that a) the door is opened every half an hour b) the working day is
6 hours long c) the refrigerator is 3/4 empty and d) half the cold
6-7
is replaced by warm saturated air every time the door is opened and
e) the warm air is cooled to 00C in half an hour, the cooling load
= 6.4 W
= 16.21 W
6-8
cool down in 2 hours, therefore the cooling load due to 5.25 litres
10.2 W
a typical primary health centre would be the sum of all the cooling
following reasons.
(3) The battery stores the extra energy available during the high
Fig 6.2 shows a block diagram of the system components. The average
COPoverall 1c4x ?1 ?
= mxCOPvc
(-tl
INSOLATION
40-
ýý=01
PHOTOVOLTAIC
PANEL
o
'IOW/m2
IA"''ý
BATTERY %=0.8
INVER-
TER ýI
=0"B4
I"A"'?-Nbl
Nm 0-57
MOTOR
I"A""'? "
ý"ýb"ý;
EVAPORATOR CONDENSER
'fr, 9
EXPANSION
VALVE
insolation (see the left hand side set of curves) and operating
temperature of the array (see the right hand side set of curves).
The curves in fig 6.3 show that the output of the array is directly
affected. The peak power of the PV array cited in fig 6.3 decreases
Each I-V curve has a maximum-power point (see fig 6.4) at which the
maximum-power-point.
clear day (i. e. clearness index > 0.7). At other times or when sky
is not clear, the chosen battery voltage would cause the array to
deliver less, but not far less, than the maximum possible power.
(-'3
NOTES. i 16 ,e
3
z
1. hnels (tkWnn2) at 25'C 14
are measured under full sun illumination
Minimum performance IS 2 watts less than W
:3C tall temperature
peak. The ruling sp. ClhCation is peak walls. For a more detailed 12 ¶2
ä
explanation, sw our Electrical ºMormancn Measurements bulletin. v 8
10 i0
2. Electriui cnwacbnstua vary with temperature.
0e 06
Voltage
dincrestift
ecreases by abo" 06 06
Current inonase$ by
Pat) dsenaaas by
25uu'Gam2
below
25'C 04 04
POS 11 by o axrc below 251C 02 02
(pack) Eacraaws by above
048 t2 16 20 24 29 04e 12 16 20 24
VOLTAGEM VOLTAGEM
NOTE: TMw eurm we npnwnbtiw of the ps(ovm. nt: I of typical pw4 N tM tMnlnS1ll, aiMO
tiny WAItInfl puipms. t such 48 OuoMs, Cabling, etc. Thee curt" we intended for reference onot,
CuroM tot the SX"100 and $*"120 5. neb we awww N from Solves Marketing.
ReUcWty and
Mechanical Specifications Environmental Sp. dfications
Optional)
_1_'I
'ti. MtI" a LI^ fuMal
` Mwnikp "I $WvcIwo
Wa_ p$ long
LDckwasM, Specifications are subject to c wq. odhoul not". 6024.1 1W
Fig 6.3 Specification sheet for a typical silicon solar panel [2]
6-14
INSOLATION CONSTANT
VOLTAGELOAD
'MAX
MAXIMUM-POWER
POINT
0-75,
MAX
L 0-401MAX
º
L
(_i º
1
1
1
º
° 071MAX
Voltage
Thus for the purpose of this analysis the efficiency of the PV array
silicon array, by
array.
on the top and bottom surfaces which are exposed to the ambient
reflected). Thus for the thermal analysis of the array, its surface,
photons have energy less than 1.1 eV (the band gap for silicon) and
insolation level.
the heat transfer rate from the array equals the energy rejected by
6-17
of the array is equal and uniform all over the surface area.
For the lower face of the array the relation used, given by Ozisik
For the upper surface of the array the relation used, given by Wong
[4], is
6.3.1.6.1
mixture of pure lead, lead dioxide, and sulphuric acid. The reaction
and forklift trucks) are heavy duty batteries which can last upto
`ýJ
competitive prices. Most of the PV refrigerators marketed employ
produced for 115/240 V ac use. Thus to cover the more general case
the circuit.
6-19
motors, due to their small size, are not very efficient. Efficiency
relation
R=1- (te
- te)/265 6.6
by a factor
in the range 0.5 to 0.8. Due to the small size of the compressors
constitute the basis for the numerical model which was used to
The final output of the model is the size of the PV array and the
refrigeration system.
and motors are variable under varying operating conditions. But the
heat, is obtained
convection.
the difference between the two values is greater than the required
& 6.8 which is then multiplied by the efficiencies of the motor, the
according to relation
Aarray 6.10
= 9frig/COPoverall/I
load of gice (in Watts) for the ice-packs. If this load is taken off
exist an optimum balance between the array area and battery capacity
The amount of daily energy required after sunset will vary with the
to meet a load of 9f rig (in Watts) through this period. Thus the
Again the 50 percent rule applies for maximum drain and therefore
the total storage capacity the batteries to meet the daily drain
will be
The PV array should meet the daily load of 9f rig (in Watts) and top
If the daily drain is less than the energy drain during three days
Thus the total array area required to meet the total system demand
will be
+ Aarray, 6.18
Aarray, total ° Aarray add
A the wind effects were not considered while calculating the heat
transfer from the array, the model would predict a higher estimate
and insolation was linear and the two were related by the expression
'1c' 6.19
1cref+0.001(Iref-I)
figures.
insolations. It was evidently clear from the graph that the COP was
F- ý
Z
W< o in 0 wt
0
LO
20-
a)
W
I --
4)
c0
d
8 a
e
a)
4)
42
a)
9 6
N
E 0
0
+1
4.3
gLL.
0
Z
co
Z u
O c)
ti-1
Z O
00 4r
w
a)
Co
L.
t.
?
0 m
c)
n)
S
N
a)
12
a
Co
c.
ü
lO
00
9.
Oa t0 NO CO t0 NO 00 ýO
2 O' C1 Of O, Of co co 0a co ON1,
00 0 0
o0 0 0
E
Co
z Co m
bO
c
v
0
cn
z y
O
[
Co 0
Ü a'
aý
M)%.
_O L.
W
Mm
MWc, 9.03
0
m 4.1 Co
W
Co
NIý No (n
L. r:
L. vl
90 Z
NWCä
+1 O
2O
Co
Q +-1 >
41
N Cli
+1 O
9. trr
N>
OO
4)
to 4.3 Co
00 N
C C1
O +)
m
O 43
.CW
a++
Co .92
'.D
%0
OG
+i
94
o0O o0 O0o
0.30
0.28
(L) 0.26
O
0.24
w
U
Q
0.22
m
Of
O 0.20
LL
Q-
0.18
1
0
0.16
E-
Z
W 0.14
U
LL-
0.12
O
U
0.10
0.08 LI. I, 1
15 20 25 30 35 40 45
AMBIENT TEMPERATURE(°C)
Fig 6.7 Variation in overall COP of PV-refrigerator with the
change in ambient temperature at various insolations
6-31
media (e. g. water) which would add more complexity to the system and
depicted by fig 6.10 for the proposed 100 watt vaccine store.
Fig 6.11 was drawn for a refrigerator of 110 W which had lesser
6-3z
INSOLATION(Wm"2)
34 100
AVERAGEMONTHLY
E 32 SUNSHINE HOURS =
30
ry
O 28
26
W
0 24
Ll- 22
W
20
18
200
0
16
r
14
012 300
LL-
10
400
LL' 8 500
»6 600
700
4
0 L rtiiIri1
15 20 25 30 35 40 45
AMBIENT TEMPERATURE(°C)
Fig 6.8 Graph of array area against ambient temperature
various insolation conditions
6-33
- 33
50
Of
0 40
ý 35
W
30
25 NM-2)
O
O
20
100
0
U-
15
*4ýý
Q 10 0,4ý*Iftwý 200
Q
300
r5 400
500
Q 600
700
56789
watts. When the figures 6.10 and 6.11 were compared, it was observed
sunshine hours upto six hours but further increase in the sunshine
required to meet (a) the daily energy demand after sunset, which
and (b) the energy required to run the refrigerator, without ice
heat gains.
6-35
0
0
61
vi
O
Oi
In
0
Cd
W
z
V)
Z
oýN
U
e 0
rW
j._ W
OMN 0
mQ 1 Q
QW
a.
m
W oJ
f-
Z
0
in
0
cri
0
.r
8N3NCN
COO t0 <D *NO2
(ud) UOA.
Vd301U332IM 001 dOd JlLI3Vdd0 A 3llV8
6-36
CD
u,
of a,
0
.i
43
.4
0
o0 0
t)
a v) : ft
c,
C13
O
0
4,
W c,
Z 0
4)
ca
L.
N a)
bo
Z
c.
U a.
a,
0 oW c.
W O 4)
Z
WD
0 0 0 O
N
0
LO 14- M r- Q 0
r-
QW ea
C-
L.
0
W oJ o,
H to S
Z 4)
0 u
M
.n a
Ln cd
u
a)
0 41
cri 43
ca
ao
uý
.r
eo
w
0
.r
I I. I. I. I
I. I. .I.
.I. -1-1
0 gN(NNN888fN8
MNf
6.3.3 Conclusions
tropics).
competitive.
This has so far been the only system which is being used by WHO to
which means large size PV-arrays but same is true of the thermal
6-38
increase the size of the array and the battery storage and increase
over the past few years. For example the price of a PV refrigerator
in 1985 was more than £4500 [10] and now in 1989 these units can be
countries.
6-38a
PART II
will be studied.
involved [11]. The aim of the analysis which will be carried out in
ammonia pair.
on the diagram.
section.
6-41
6-42
bars. The manufacturer quoted a COP was 0.23. The analysis of the
ammonia in the absorber can be read against this point which is 3.5
the point E on the pure ammonia line at a pressure of 4.5 bars will
flowing towards the absorber through the annulus, and the warm
tube, towards the evaporator. This way the heat is exchanged between
SI71
xrý,`
#M
%, y0,ß
xrrr `..
if
N
to
0
.o
a :::
'3
::: i: LfN:
of iý s°, o^
I ýo i. a 4- Ia
6-44
achieved with simple flat plate collectors. Hence the system needs
The flow of weak solution from the generator to the absorber per,
Y= (1 - kr)/(kr - kW)
kg of ammonia generated.
[13]
(1 - vrg)(28.7 + 9xvwg)
Qgas = ------ -_---- xtT kJ/kg NH3 6.19
17(vrg - vwg)
After consultation with Green [16] the value for the difference in
partial pressure was assumed as 0.2 bar. Thus the partial pressures
125°C.
the absorber will be around 0.7 bars which indicate that the partial
evaporator exit was calculated as 10.2 bars. This means that the
produced. Thus it can be said that the modifications carried out are
should evaporate completely below 0°C. This means that rich mixture
then be possible.
Now the bubble pump duty can be calculated. The amount of weak
about five times what the pump was circulating under standard
6.4.4 Conclusion
in fig 6.14. The high temperature flat plate collector replaced the
L.
0
4)
U
d
ri
0
v
r-1
co
0
N
a)
a
ca
C) 0
u" /CC
4)
0
r oc
v w 0
ä
ný N
d v N
C)
6.
w
r
v
m 4.4
a)
3
ca
L.
I- p
d
ob
L c
aý v
C u
a, x
V I- a)
C a)
8 4-
a
0 a, w
a L
Q ,o
12 0
.' Ei
Q ca
J w
0
C)
.i4.3
co
I- c a)
0 aý
v C)
ä u
v 0)
r r.
W 0
ao
0
lD
6-51
heat is supplied to the generator. Thus for the period when there is
components.
6-51a
PART III
6.15. The chosen refrigerant defines the saturation line, while the
temperature fixes the pressure, Pe, for the chosen refrigerant and
a
I-
Y
v
W
0
U-
C)
J
u
0
a-
ü
w
ID
w
Z
F-Z
au
38nSS3addO 901
6-54
temperature.
temperature,, is given by
DG = RTln[p/ps(T)] 6.20
equilibrium)
m m0exp[D/E]n 6.21
ln[ps(Te)/ps(Ta)]
T2 Ta 6.23
1n[ps(Tc)/ps(T2)]
wrtk
Thus the above equation 6.23 provides oneAa method to determine the
minimum generation temperature, T2, when one can define the function
In the next section, this new found basis for the suitability of a
appendix B). This would mean a day time temperature of 43°C and a
(shown in fig 6.16 and 6.17 respectively) were prepared which depict
at -100C.
limited to 120°C).
6-57
0
c
co (0
I
a)
e
o'
0
c.
U c°
c cý
a
oe
W
4)
W
0
m boa
W
00
U
II *Z
O
c+'
0
co
+
z bo
\to Z>ý
co
w w
°f D G,
L
. Z .cc o
a'
O nom
ca
U V
sö
w o ar
02 co
äC
(0
'')
0
c. c
t7 +ý
v
M
n bo
ar
N8°
(00) 32jl1V3dW31 NOIIV N30
6-58
Co
In
IC
O
Nw
O
Ul)
cli
`/
d
U 43
0O
II Ld L. d
W CL 4) rl
to
2 4) >1
00 V
z Q
W
F 4r C
OO
I-
W co
z o
W
F- D Cl) 4.4 vi
L. L.
w Z
cd 44
O W
gý
40
U .4
w ty-
O cä
ö
a
0
410
m cu
CD
ý
.cc
aco
co
c. C
bo
+14
994
NO N.
Op
110°C.
Using the graphs presented in fig 6.16 and fig 6.17 one can see
Some other more common refrigerants were also examined on the basis
are also listed in the table and one can see that all but sulphur
TABLE 6.1
vapour sorption units. Table 6.2 compares the pressure and specific
The final choice between the two cannot be made in isolation. For
sorbent, its performance with the particular sorbent and the design
of the collector-cum-generator.
6-62
TABLE 6.2
(estimated from the value at normal boiling point, using the Watson
relation [22]) which means that, ignoring the sensible heat of the
absorbent and the equipment, the pair can show maximum COP of 0.74.
This is a very reasonable figure but one has to see if the reaction
moles of methanol had been reported desorbed after 4 hours when the
heating fluid was at 110°C. These temperatures are within the limits
reasonable. But one has to look into the sorption process as well.
pressure of the salt mixture must be more than 23 mbar. The reaction
temperature of 37.7°C.
would not take place. Thus, it was concluded that the calcium
the limited available sunshine period the kinetics of the system has
sky radiation was not enough to achieve the desired rate of reaction
The heats of reaction Q, and Q2 are about equal and vary between
2200 and 2500 kJ/kg NH3 [231, which is approximately twice the
side and indicate that the thermodynamic cycle COP of this pair can
Solution to this problem had been found [23,24] by adding some other
monovariant and one does not have the freedom of choosing the
lI
ö
=üZ Co
a--
N
co
O 0
S
V
vi
NZ a,
O
Vi 0
Ö 0
O B
W
u
cd
U
L
co
a
Ou m
o
0
N w w
cd
LU 00
V ca
V
0
W
t 2- a)
a
Z ca
N LU U
l"'
co
40
Csr
0
NO co %0 N V- CO %0 _t N
--t
JDG 601)
(BIMS 38nSS38d
6-68
requirements [25]. The price of US$ 6,000 is very high and reflects
adsorbents.
the pairs.
6-69
following conditions.
generation temperatures below which the cycle would not perform. The
COP at these temperatures was zero. It can be observed that all the
The COP so calculated did not take into account the energy which
&-70
A.C.-208C/METHANOL
0.44
0.42
0.40
0.38
0.36
w
z 0.34
< 0.32
0.30
O 0.28
X 0.26
a-
0.24
0.22
0.20
z
0.18
w 0.16
0.14
li.
w 0.12
O 0.10
O
0.08
0.06
0.04
0.02
0.00
110 115 120 125 130 135 140 145
GENERATIONTEMPERATURE(°C)
Fig 6.19 COP of various adsorption pairs in an intermittent cycle over
a range of generation temperatures (water-cooled condenser)
6-71
6-71
A.C.-208C/METHANOL
0.38
0.36
0.34
0.32
Ü
0.30
Q
0.28
0.26
C 0.24
w 0.22
a-
0.20
I
0.18
Z 0.16
W 0.14
U
0.12
LL-
W
0.10
0
0.08
0.06
0.04
0.02
0.00
120 125 130 135 140 145
GENERATIONTEMPERATURE(°C)
Fig 6.20 COP of various adsorption pairs in an intermittent cycle over
a range of generation temperatures (air-cooled condenser)
6-72
system. This means that the COP of ammonia system would lowered in
summarized as below.
months.
(iii) The storage batteries have a normal life of 2-3 years. This
maintenance.
6-74
But in part II of the chapter it was found that to adapt the system
have called for a new manufacturing line, which for such a limited
solid sorption refrigerator. This was dealt with in part III of this
for such a plant. It was further established that under the WHO
temperature under 320C night time ambient temperature and was not in
the race any more. Calcium chloride ammonia systems were found to
table 6.3). This would slow down the response of the system under
transient conditions and make the heat transfer in the sorbent bed
adsorbents.
6-76
TABLE 6.3
(a) activated carbon has the lowest specific heat and the highest
(see table 6.3). This means that the response time, under
requirements.
REFERENCES
1977.
NASA TN D-5881,1970.
pp 1-12,1985.
August 1989.
1985.
U. K.
6-80
pp 51-120,1966.
'A heat pump and thermal storage system for solar heating and
102, pp 59-65,1980.
no 5, pp 447-455,1985.
August 1989.