SIMULATION OF A SOLAR POWERED ABSORPTION REFRIGERATION SYSTEM

Kamal Nasr-El-Din Abdalla * and Fath-El-Rahman Ahmed El-Mahi * ABSTRACT Many parameters affect the performance of solar refrigeration systems. This paper presents theoretical analysis of a solar absorption refrigeration system, which use aqua-ammonia solution in an intermittent cycle. A computer program has been developed. This program is useful in studying the performance of any intermittent absorption system at different conditions of operation. The simulation program was used to predict the performance of an existing system and good agreement was obtained. Keywords: solar refrigeration, intermittent absorption, Aqua-Ammonia 1. INTRODUCTION
Condenser Generator

An absorption refrigeration system uses heat as a source of power, instead of mechanical energy,[1]. Figure 1, shows the basic absorption . refrigeration system, which consists of two cycles: 1-The regeneration cycle, during which, heat is supplied to the working solution, contained in a generator, to drive the refrigerant off the solution, which is then condensed and stored as a liquid in a receiver. 2- The absorption cycle, during which, the refrigerant vaporizes, and hence, produces cooling effect and flows back to be absorbed by the weak solution, contained in an absorber. In continuous absorption systems, the regeneration and absorption cycles, take place at the same time. In intermittent systems, the absorption cycle, takes place after the regeneration cycle has been completed [2]. Figure 2 shows a simple intermittent absorption cycle. Solar energy is suitable with intermittent absorption refrigeration systems, because, solar energy, itself, is intermittent in nature, [3]. Solar intermittent absorption refrigeration systems have been studied by Chinnappa[4], and Swartman and Swaminathan[5]. In their

Throttling Valve

TV

Evaporator Evaporato

Absorber

Pump

Figure 1: Absorption Refrigeration Cycle

qg

qc

a- The regeneration cycle

qa

qe

b- The absorption cycle Figure 2: Simple intermittent absorption cycle

* Faculty of Engineering and Architecture, University Of Khartoum ** Faculty of Engineering Technology, Nile Valley University

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

experiments, the generator was an integral part of a flat plate collector and they used separate vessels for the generator and the absorber. In this study there is a single vessel performing both functions, of the generator; during the regeneration cycle; and the absorber, during the absorption cycle.

S = I b R b ( ) b + Id ( )d (1 + cos ) / 2 + g (I b + I b )( )g (1 cos ) / 2 (1)

Where, I b and I d are the beam and diffuse radiation, and are given by the expressions: I b = b .Io Id = d Io (2) (3)

2.

SYSTEM DESCRIPTION

Figure 3 shows a schematic diagram of the system used in this study. During the regeneration cycle, water is heated in a flat plate collector and allowed to enter the generator, through a coiled tube which is immersed in aqua-ammonia solution in the generator. As a result ammonia vapor rises up and passes to a water cooled condenser, where it condenses and flows by gravity, to be stored as ammonia liquid in a receiver. During the absorption cycle, cooling water is allowed to enter to the generator, which now acts as an absorber, through the same tube coil, as a result, the temperature and pressure of the weak solution reduce and, hence, ammonia liquid vaporizes in the receiver, causing cooling effect there and flows back, to be absorbed by the weak solution in the absorber.

I o is the solar radiation incident on a horizontal surface, outside of the atmosphere, for an hour period, and is given by;

Io = (12 3600 / )Gsc{1 + 0.033cos(360n / 365)} {cos cos (sin 2 sin 1 ) + ((2 1 ) sin sin ) / 180} (4 )

3.2 Generator analysis The actual heat received by the solution in the generator, is given by;

Q u = rm .{S 3600U L (Tpm Ta )}

( 5)

Where rm = A m /A c is the model ratio. Tpm is the mean absorber plate temperature. Ta is the ambient temperature. U L is the overall heat loss coefficient.

3.

MATHEMATICAL MODELING

3.1 The flat plate collector The total radiation absorbed by the absorber plate of the collector during one hour, is given by the expression, [3]:

Cooling water

condenser

Solar collector

Receiver/Evaporator Generator/Absorbe

Figure 3: A schematic diagram of the refrigeration system 2

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

At the initial state the following terms are defined: Ts1 = initial temperature of the solution, x l1 = initial concentration of ammonia in the solution, h l1 = initial enthalpy of the solution, Pl1 = initial pressure of the solution. After one hour, a certain quantity of heat is transferred to the solution and the new state of the solution is defined by: Ts2 , x l2 , h l2 and Pl2 . If, after one hour of heating, the solution temperature, Ts2 , is less than the saturation temperature, then, no ammonia vapor will be generated and the whole energy, Q u , is used to raise the solution temperature from Ts1 to

fraction is also used to generate ammonia vapor. The energy balance equation for this case is, therefore: m s cp s (T s2 T sat ) + m v h v = r m .(1 U r ) {S 3600U
L

(T sat + T s2 )/2 T a }} (10)

Where m v is the mass of the ammonia vapor generated during the period (1 Ur ) , and h v , is the enthalpy of the generated vapor. From equation (10), the temperature of the solution at the end of the hour can be found as: Ts2 = {rm .(1 U r ){S 1800U L (Tsat 2 Ta )} +
m s cp s Tsat m v h v }/{m s cp s + 1800(1 U r )U L rm } ( 11)

Ts2 , and the energy balance equation is:

m s cp s (T s2 T s1 ) = r m .{S 3600U From which,
Ts2 = rm .{S + (( m s cp s /rm ) 1800 U L )Ts1 + 3600 U L Ta }/ (m s cp s + 1800 U L rm ) (7)
L

(T s2 + T s1 )/2 T a } (6)

If Ts2 , is greater than the saturation temperature, then, only a fraction of the useful energy available during one hour, is used to raise the temperature of the solution to Ts2 . Since the rate of energy is assumed to be uniform throughout each hour, this fraction also gives the period of time required, for the solution temperature, to reach the saturation temperature. Let U r to represent this period, equation (6), then, becomes: m s cps (Tsat Ts1 ) = rm .U r {S 3600U L
{(Tsat + Ts1 )/2 Ta }} ( 8)

Knowing the temperature Ts2 ,the properties of the aqua-ammonia solution, and the evaporated ammonia, at the end of the hour, can be found, at the pressure Psat . These include the concentration of ammonia in the solution, x l2 , the enthalpy of the solution, h l2 , the concentration of the ammonia vapor, x v2 and, the enthalpy of the vapor, h v2 . The mass of the generated ammonia vapor during the given period, is: m v2 = ms1 (x l1 x l2 )/(x v2 x l2 ) ( 12)

And the mass of the remaining solution, at the end of the hour, is:
m s2 = m s1 (x v2 x l1 )/(x v2 x l2 ) ( 13)

The amount of water vapor contained in the evaporated ammonia is given by:
m w2 =m v2 (1 x v2 ) ( 14)

From which: rm {S - 3600UL {(Tsat + Ts1 ) / 2 Ta }} ( 9) For the rest of the hour, the remaining fraction of energy, (1 Ur )S, is used to raise the solution temperature from Tsat to Ts2 at Ur = mscps (Tsat Ts1 ) /

If the final temperature, Ts2 , of the solution at the end of an hour, as obtained from equation (7), equals the saturation temperature of the solution, Tsat , then, for the next hour, the whole available energy, is used to generate ammonia vapor, and to raise the temperature of the remaining solution. In this case, the energy balance equation, and the
3

constant pressure Psat . At the same time, this

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

final temperature of the solution, are given by equations (10) & (11), respectively, with U r =0.
3.3 Condenser analysis

absorption. Considering one hour duration, the energy balance equation, is therefore, [1]: m s1 h s1 + m av h g = h f1 m s2 (18)

The quantity of heat which must be removed, from the generated ammonia vapor in the condenser, during one hour is given by:

where ms1 and h s1 , are the mass and the enthalpy of the solution at the beginning of the hour, h g and m av are the enthalpy and the mass of ammonia vapor, which is absorbed during one hour of absorption and ms2 and h f1 are the mass and enthalpy of the solution, at the end of that hour. From equation(18), h f1 can be found as:

Q hr = m v (h v2 h cd )

(15)

Where, m v , is the quantity of ammonia vapor which is generated during that hour, h v2 , is the enthalpy of the generated ammonia vapor, and, h cd , is the enthalpy of the condensed ammonia liquid. The quantity of heat absorbed by the cold water in the condenser, during one hour, must be the same as the quantity of heat rejected by the ammonia vapor during that hour, so the energy balance equation in the condenser is: m cw cp w Tcw = m v (h v2 h cd ) (16)

h f1 = h s1 + m av (h g h s1 )/m s2

(19)

where m cw is the total mass of the cooling water required in the condenser, cp w is the specific heat of the cooling water and Tcw is the rise in temperature, of the cold water, after one hour. From equation(16), the mass of the cooling water required in the condenser is: m cw = m v (h v2 h cd ) / cp w Tcw (17)

Equation (19) gives the enthalpy that the weak solution would have after one hour of absorption, if there was no cooling. Due to cooling, the enthalpy of the solution will be reduced to a value h f2 , corresponding to the absorption temperature Tabs , and the new value of concentration of the solution x s2 , which is given by:
x s2 = x s1 + m av (x
v

x s1 )/m

s2

(20)

Where, x s1 , is initial concentration of the solution, and, x v , is the concentration of ammonia vapor. Hence the heat of absorption , is: H = ms2 (h f1 h f2 ) (21)

3.4 Absorber analysis After the regeneration cycle has been completed, there will be a certain quantity of ammonia m vt evaporated, condensed and stored in the receiver at a known condenser pressure Pcd and temperature Tcd . Due to cooling, the pressure of the weak solution in the absorber, will be less than the pressure of the ammonia liquid in the evaporator. Hence ammonia liquid starts to evaporate in the evaporator, produces cooling effect there, and flows back to be absorbed by the weak solution, in the absorber. Due to this process , a quantity of heat will be released in the absorber, this is known as the heat of 4

Equation (21) gives the quantity of heat which must be removed by the cooling water during one hour of absorption. Hence the mass flow rate of the cooling water during absorption is given by:
m cwa = m s2 (h f1 h f2 )/3600cp w Tcwa

(22)

Where, cp w , is the specific heat of the cooling water, and Tcwa , is the rise in the cooling water temperature. The pressure of the solution at the end of one hour of absorption, Ps , is found from the h-x chart, which is prepared by G.G. Brown[1].

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

3.5 Evaporator analysis The quantity of heat absorbed by the evaporated ammonia, in the evaporator, during one hour, equals the quantity of heat removed from pure ammonia liquid in the receiver, and from the water surrounding the receiver, during the same hour. The energy balance equation is therefore:

mavh fg = cpa mca (Tc1 Tc2 ) + cpw meq (Tc1 Tc2 ) (23)

Initially, the aqua-ammonia solution, is at state(1), which corresponds to a concentration of 41%, a temperature of 30C and a pressure of 2.3 bar. When heat is supplied, the temperature and the pressure of the solution increases until the saturation pressure is reached. Below the saturation pressure, the whole quantity of heat, which is supplied to the generator, is used to raise the solution temperature. No ammonia vapor is generated, and the concentration of the solution, remains constant at 41%. When the saturation pressure is reached, ammonia vapor starts to rise up from the solution, at constant pressure, the concentration starts to decrease and the temperature continues to increase, until the regeneration cycle is completed. During the absorption cycle, cooling water is used, to cool down the aqua-ammonia solution in the absorber, to 20C. Ammonia liquid starts to evaporate in the evaporator, and flows back to be absorbed by the weak solution in the absorber. As a result, the concentration and pressure of the weak solution increase. This continues until the whole ammonia liquid is returned back to the absorber. Comparing the results for the three cases of condenser temperature, it is clear that, the input energy is high, at higher condenser temperature, and the coefficient of performance increases as the condenser temperature, is reduced. This means that, to obtain a high coefficient of performance, then the condenser cooling water, must be provided at a lower temperature, which means additional cost to the system. In solar refrigeration systems, it is usually required to design the system, so that the overall cost is minimum, and, since the variation in the refrigeration effect is small for the three cases of condenser temperature, it may be desirable, to design a system, with a lower coefficient of performance, if the cost is significantly reduced. On the other hand, if the condenser temperature is high, then there will be no additional cost for provision of cooling water, but the working pressure will be high,
5

where Tc1 & Tc2 are the initial and final temperatures, in the receiver, cp a & cp w , are specific heats of pure ammonia, and cold water, respectively, m eq , is the equivalent mass of the cold water, accounting for the metal in the evaporator. m ca , is the total mass of the remaining liquid of pure ammonia in the evaporator. h fg , is the latent heat of evaporation of pure ammonia, and m av , is the mass of ammonia vapor, which is transferred from the evaporator to the absorber during one hour. The temperature in the evaporator after one hour is therefore:
T c2 = T c1 m (m
ca av a

h fg / + m
eq

.cp

.cp

(24)

3.6 The simulation program Using the equations derived, in the previous sections, for individual components, a simulation program was written for a solar refrigeration system, which uses ammonia as a refrigerant. The simulation program is written in the Fortran programming language. The least square method of curve fitting, [6], is used in the program, to evaluate the properties of the fluids involved. The user is asked to enter data concerning: size, site, date, available cooling water temperature and the required absorption temperature.

4.

RESULTS AND DISCUSSION

The following table shows a summary of the output of the program, for condenser temperatures of 25, 20 and 15C. These results are plotted in Figures, 4,5 and 6, respectively.

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

and hence, the system must be designed to withstand this high pressure. Swartman and Swaminathan,[5], experimentally studied the operation of an intermittent absorption refrigeration system, using aqua-ammonia solution, and operated by a flat plate collector. In their system, the generator and the absorber were separate vessels, in addition the, the solar collector was part of the system generator. In their

experiment, the coefficient of performance, was found to be about 0.06, at a condenser temperature of, about, 30C, [1]. Figures 4, 5 and 6, also, show comparison of the theoretical predictions, which are obtained in this study, with experimental work, which was carried out on a refrigeration system model, for the three cases of condenser temperature. Table 2 gives a summary of this experimental work.

Table1: Summary of the output of the simulation program

25 Condenser Temperature (C) Duration (hours) Total mass of the solution (kg) Initial concentration of the solution (%) Initial temperature in the generator (C) Regeneration Saturation temperature (C) Cycle Maximum temperature in the generator (C) Initial pressure in the generator (bar) Saturation pressure (bar) Maximum pressure in the generator (bar) Total energy received(kJ) Quantity of the generated ammonia vapor(kg) Duration (hours) Absorption temperature (C) Absorption Initial evaporator temperature (C) Cycle Minimum evaporator temperature (C) Refrigeration effect (kJ) Coefficient of performance (cop) 7 4.1 41 30 80 84 2.3 10.1 10.1 2092 0.167 2 15 30 9.4 129.2 0.062 20 6 4.1 41 30 74 78.6 2.3 8.57 8.57 1847 0.187 2 15 30 7.1 143.6 0.078 15 5 4.1 41 30 67 73.4 2.3 7.4 7.4 1543 0.19 2 15 30 6.9 144.8 0.094

Table 2: Summary of test results

Regeneration cycle

Absorption Cycle

Test number Duration (hours) Condenser cooling water temperature(C) Maximum solution temperature (C) Maximum solution pressure (bar) Duration (hours) Absorber cooling water temperature (C) Minimum evaporator temperature (C) Actual coefficient of performance Theoretical coefficient of performance

1 7 18-27 93 12.5 2 16-23 15 0.019 0.062

2 6 16-24 86 10.5 2 17-22 16 0.02 0.078

3 5 11-18 82 9.4 2 17-23 15 0.026 0.094

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

30 120

35

40

45

50 120

1 5.5 4
100

b ar
3 6 2

Temperature C

80 60

1 1.6 6 7

Theoretical cycle Actual cycle 100

80 60

2.9 1
40 20 0 30

1. 9 1.1 9 48

40

1 9 5
40 45

20 0 50

35

Concentration % Figure 4: Actual and theoretical cycles for Condenser temperaure of 25 C

30 120 100 35 40 45 50 120

1 1.6 6

Temperature C

8. 57

b ar
7 3 6 2

Theoretical cycle Actual cycle 100

80 60 40

80 60 40 20 0 30

2.9 1
1. 9 1.1 9
8 4
35 40

1 9 5
45

20 0 50

Concentration % Figure 5: Actual and theoretical cycle for Condenser temperature of 20 C

30 120 100

35

40

45

Temperature C

8.5 7 b ar 6.1 5 2.9 1 1. 9 1.1 9

4
35 40

Theoretical cycle Actual cycle 100

3 7 6 2
80 60 40

50 120

80 60 40 20 0 30

1 8 9 5
45 20 0 50

Concentration % Figure 6: Actual and theoretical cycle for Condenser temperature of 15 C

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44
7

5. CONCLUSION In this study, a computer simulation program was written, using the Fortran language, to predict theoretical performance of an intermittent absorption refrigeration system. The program was run for different values of performance parameters, and gave good results. The program calculates the properties of the fluids encountered in the experiment, and gives values for different required quantities. These include the quantity of the cooling water required in the condenser tank, the required length of the condenser coil,etc. Finally, the program gives the value of the theoretical coefficient of performance of the system, according to the given input data. The results obtained from the program showed that the coefficient of performance of an intermittent absorption refrigeration system is high when the condenser temperature is low. The theoretical results, obtained from the program, were compared with actual performance of an existing apparatus, and it is found that, the theoretical performance is higher than the actual one. The difference between the two results is, largely, due to inaccuracy in measurements, which were taken during the experiment.
It can be concluded that, for intermittent absorption refrigeration systems, the energy supplied to the generator of the system is high, when the temperature of the condenser is high, and that the coefficient of performance is high, when the condenser temperature is low. In solar refrigeration systems, it is desirable to design a system which can utilize solar energy, at low cost. This study showed that the variation in the refrigeration effect, for the three cases of condenser temperature, is small. It seems that, it is better to design a solar intermittent absorption refrigeration system, which uses tap water, for condenser cooling. This may results in a low coefficient of performance,
8

but its main advantage is to avoid the additional cost of providing cooling water at a low temperature.
Nomenclature A surface area (m) Ac collector area (m)

Am cop cp G sc h h h fg m n P
p sat

model area (m) coefficient of performance specific heat (kJ/kgC) solar constant (W/m) heat transfer coefficient (W/mC) specific enthalpy (kJ/kg) latent heat of vaporization (kJ/kg) mass (kg) day of year, index of refraction pressure (bar) saturation pressure (bar) quantity of heat (J) absorbed solar radiation (kJ) temperature (C) saturation temperature (C) collector overall heat loss, coefficient (W/mC) concentration of ammonia in aquaammonia solution

Q, q S T Tsat UL x

Greek: radiation absorptance slope angle, (degrees) surface azimuth angle (degrees) declination angle (degrees) angle of incidence (degrees) z zenith angle (degrees) radiation reflectance radiation transmittance latitude angle (degrees) hour angle (degrees)
Subscripts:

a b c d g l m s v

ambient beam collector diffuse ground liquid model solution vapor

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44

REFRENCES 1. McQuiston F. C. and Parker J. D. , Heating, Ventilating and Air conditioning, analysis and design- 2nd edition John Wiley& Sons, (1982). 2. Carl H.Turnquist/Alfred Bracciano/ Andrew Daniel Althouse-Goodheart, Modern Refrigeration and air conditioning, Willcox , Publisher, (2000). 3. Duffie J. A. and Beckman W. A., Solar engineering of thermal process- 2nd edition-Wiley-Interscience, (1991).

4. Chinnappa J.C.V. Performance of an intermittent refrigerators operated by a flat-plate collector Solar Energy, volume 6, PP 143-150 (1962). 5. Swartman, R.K. and C.Swaminathan Further Studies On Solar-Powered Intermittent Absorption Refrigeration, Paper at the International Solar Energy Society Conference, 6/114,7 pages, Melbourne(1970). 6. Danilina N. I., Dubrovskaya N. S., Kvasha O. P., and Smirnov G.L., Computational mathematics, Mir Publishers, Moscow,(1988).

Sudan Engineering Society JOURNAL, August 2005, Volume 51 No.44