AU J.T. 10(1): 45-51 (Jul. 2006)

An Experimental Study of Hunting in Evaporators C. O. Adegoke and M. A. Akintunde Department of Mechanical Engineering, Federal University of Technology Akure, Nigeria

Abstract It had been reported by many researchers that evaporators exhibit some trend of hunting in the early part of superheat. Many equations predicting the actual behavior had been presented but they have not been able to minimize the effect of the evaporator hunting. This work investigates the minimization of the hunting behavior of evaporators at the onset of superheat using balanced point approach. A design model based on balanced points between the operational components of vapor compression refrigeration systems (Akintunde, 2003) was used to interpret the result obtained from the constructed rig. In addition, test data have been used to assess the quality of the computer simulation results. The simulated and the experimental performance results were compared and the results obtained from the experimental investigations justified adequately the developed design. From the experimental investigations it could be concluded that, the balancing of the operational components enhanced the performance of the evaporator. Maximum absolute deviations of the rig parameters from the model are within the range of 16 % to 19 %. Keywords: performance.

Hunting,

computer

simulation,

balanced

points,

evaporator,

represented as linear elements. Najork (1997), whose analysis was based on linear valve characteristics, referred to the possible nonlinear (or hysteresis) behavior of the valve. In their simulation, Yasuda, et al. (1983) used a semi-distributed model for the evaporator and a linear model for the TEV. The oscillatory behavior of the system was attributed to the changes in the heat transfer coefficients on the refrigerant side of the evaporator, which were presumably caused by random fluctuations of the transition plane. Along the same line of reasoning, Wedekind (1997) suggested that random fluctuations in mixture-vapor transition plane could cause fluctuations in the refrigerant temperature. This was experimentally investigated and confirmed by Mithraratne and Wijeysundera (2003). Ibrahim (2001) looked at this problem from the energy transfer angle. This was achieved by varying the inlet and outlet conditions of the evaporator. He concluded that when a system reached stability again, the superheat and the bulb temperatures converged at higher values than the initial one. However,

Introduction In order to control the amount of refrigerants that enters into the evaporator of a vapour compression refrigeration system, many expansion devices could be employed. Some of these are: thermostatic expansion valves (TEVs), solenoid valves, high-or low-side float valves, capillary tubes, or discharge bypass. In the use of any form of expansion device it has been noticed that evaporators exhibit an undesirable behavior known as “hunting” under certain operating conditions (Mithraratne and Wijeysundera 2002). The systems variables such as the refrigerant flow rate, evaporator pressure, and superheat temperature oscillate in a sustained manner when hunting takes place. Dynamic characteristics of TEVs and stability analysis including hunting of TEVcontrolled evaporators have been the subject of many publications. Stoecker (1996), Najork (1997); Wedekind (1997) and Ibrahim (2001) investigated the stability of a TEV evaporator control loop using lumped analytical models where the evaporator and the TEV were 45

AU J.T. 10(1): 45-51 (Jul. 2006)

behavior was attributed to two factors. These are: (i) the random fluctuations of the mixturevapor transition plane, and (ii) the overall dynamic characteristics of the system. The above factors have been mentioned in a number of studies and many equations predicting the actual behavior presented the effects of the evaporator hunting minimization. This work investigates the minimization of the hunting behavior of evaporators at the onset of superheat using balanced point approach.

since the bulb and the evaporator temperatures were controlled parameters of the TEV-outlet mass flow rate, it was suggested that the instability of the latter may then be mainly due to the evaporator temperature instability. He pointed out further that the instability following the sudden decrease in the chilled water temperature vanished almost after 120 seconds of operation and the final stable values were lower than the initial ones. From the review done by Mithraratne et al. (2000), based on the results of series of experimental studies, the temperature of the tube wall where the TEV bulb is attached fluctuates if the degree of superheat is below a certain value. Many researchers had worked on this same effect using capillary tube and other expansion devices with evaporators. These can be found in the works of Meyer and Dunn (1998), Motta et al. (2002), Kim et al. (2002), Wolf and Pate (2001), Wijaya (1992), and Wei et al. (2003). The results of their theoretical and experimental investigations were similar to those of TEV-controlled evaporators. As pointed out by Ibrahim (2001), with zero superheat, the evaporator may be flooded and liquid refrigerant might pass to the compressor if no protection is used and may result in compressor damage. Hence, a certain degree of superheat between the evaporator and the compressor may be beneficial, but this actually leads to the instability of the evaporator performance. Such instability results in unstable cooling capacity and fluctuation of evaporator temperature for a certain period of time, which may have a direct effect on the refrigerated products. A scrutiny of the literature shows that the hunting of the evaporator in relation to the expansion valve is not actually the problem of the expansion valve used. This is because the behavior of the evaporator with various expansion devices is similar. On the other hand, the hunting is limited to certain degrees of superheat (Ibrahim 2001, Mithraratne, et al., 2000). Also, this superheat causes the hunting and could not be avoided for the protection of the compressor and the life of the system (Ibrahim 2001). According to Mithraratne and Wijeysundera (2002), the problem of hunting

Material and Methods Experimental Rig The experimental rig was designed using Ref-2003. Ref-2003 is a software developed by Akintunde (2003). The mathematical model used in the development of the software was based on the balanced points between the operational elements of any vapor compression refrigeration system. Information about the development of Ref-2003 can be obtained from Akintunde (2003) and Akintunde (2004). The experimental rig is a simple refrigeration unit, shown in Fig. 1. It is a 10 kW cooling system with a double tube evaporator. The test evaporator is made up of 10 tube-in-tube sections, each of length 1.85 m, arranged in a serpentine manner in a horizontal plane. The refrigerant flows in the inner tube of the evaporator, while the secondary fluid (water in this case) flows in the opposite direction in the annulus. The inner diameter of the inner tube is 5.588 mm and the inner diameter of the outer tube is 8.350 mm. The evaporator temperature was regulated (or varied) by using the Evaporator Regulator Pressure Valve (ERPV)

The Experiment According to Mithraratne and Wijeysundera (2002) and Wijeysundera, et al. (2000), the hunting of the evaporator lies between 1o and 5oC of superheat. Hence, in this experiment, the degree of superheat within this 46

AU J.T. 10(1): 45-51 (Jul. 2006)

Condenser Compressor

Expansion device

EPRV Evaporator Temperature Transducer Pump

Water tank

Heater

Qe P COPsystem

COP =

range was specifically focused on. Also, measurements were made beyond this range to justify the stability at the later stage of superheat. The following parameters were measured during the experiment: inlet and outlet evaporator temperatures, inlet and outlet circulating water temperatures and the mass flow rate while the condensing temperature was kept constant at the designed value of 40oC and the ambient temperature was assumed to be constant at 35oC.

η=

(4)

(5) COPCarnot while the subscripts are: ee evaporator exit evaporator inlet ce condenser exit condenser inlet cpe compressor cpi compressor inlet r refrigerant.

Results and Discussions

ei ci exit

As the degree of superheating increases, refrigerating effect decreases, Fig. 2. This is easily explained from the mass flow rate point of view. As the degree of superheat increases, the rate at which the refrigerant in the evaporator flashes into gas increases. This thereby increases the volume of gas in the evaporator since the condensing temperature is kept constant, thus resulting into a decrease in the mass flow rate (Lee, et al. 2002) (also Fig. 3). Normally, if saturated refrigerant leaves the evaporator it will be superheated at the end of the compression process, hence if the refrigerant leaving the evaporator is superheated, then the degree of superheat at the end of the compression process is increased. This means that more heat transfer area will be required for the condenser for an effective rejection of the heat absorbed by the

The following parameters: rate of heat absorption (Qe), rate of heat rejection Qc), coefficient of performance (COP) and compression power (P) were evaluated with the aid of the measured parameters, steam table (published by Stoecker and Jones (1982) and equations (1) to (4) suggested by Jabardo, et al. (2002). The efficiency was calculated using equation (5) suggested by Jung, et al. (1999). The measured and the calculated data from the rig were compared with those of Ref-2003 generated data. Qe = mr (hee − hei ) Qc = mr (hce − hci ) P = m r (hepe − hepi )

Fig. 1. Schematic diagram of the refrigeration system

(1) (2) (3) 1

AU J.T. 10(1): 45-51 (Jul. 2006)

refrigerant. Since the heat transfer area is fixed in this case, the rate of heat rejection decreases with the increase in the degree of superheat. This is shown in Fig. 4. Generally, the rig shows almost the same pattern for both refrigerating effect and rate of heat rejection (Figs. 2 and 4). As could be expected, the rate of heat rejection is higher than the refrigeration effect. Also, from Figs. 2 and 4, it can be seen that the characteristic curves of the rig is closer to that of the model. On the average, the parametric value of the rig is 16.89 % below that of the model. Although some fluctuations could be observed in Figs. 2 and 4 between 1o and 3oC

of superheat, this is not as pronounced when compared with the work of other researchers (Mithraratne and Wijeysundera 2001, Wedekind 1997, Ibrahim 2001; Wijeysundera, et al. 2000). This showed the effect of balancing the system components. Also, variation in the mode of heat transfer may result from the variation in the atmospheric condition (ambient temperature) and this may cause slight anomalies (Ibrahim 2001). In Fig. 5, the efficiencies of the model and the rig show similar trend. From Figs. 2 to 6, it was observed that, as the degree of superheat increases, the mass flow rate decreases and so also all the other

Qer = Refrigerating Effect of the Rig. Qem = Refrigerating Effect of the model

120 100 80 60 40

Qer

Qem

20 0 0

1

2

3

4

5

6

7

8

9

10

11

12

Degree of Superheat ( Fig. 2 Variation of Refrigerating Effects w ith Degree of Superheat

1.6 1.4 Mass flow Rate (kg/s)

Refrigerating Effects (kW)

140

1.2 1 0.8 m

0.6

Linear (m)

0.4 0.2 0 0

1

2

3

4

5

6

7

8

9

Degree of Superheat ( C) Fig. 3 Variation of Mass Flow Rate w ith Degree of Superheat

48

10 11 12

C)

AU J.T. 10(1): 45-51 (Jul. 2006)

Heat Rejection Rate (kW

30 Qcr = Heat rejection rate for the rig Qcm = Heat rejection rate for the model

25 20 15 10 5

Qcr

Qcm

0 0

1

2

3

4

5

6

7

8

9

10

11

12

Degree of Superheat ( C) Fig. 4 Variation of Heat rejection Rate w ith Degree of Superheat

0.8 0.7 0.6 Efficiency (%)

0.5 0.4 Nr

Nm

0.3 0.2 0.1 0 0

1

2

3

4

5

6

7

8

9

10

11

12

Degree of Superheat ( C) Fig. 5 Variation of Efficiency w ith Degree of Superheat

5 4.5 4

COP

3.5 3 2.5 2 1.5 1

COPr

COPm

0.5 0 0

1

2

3

4

5

6

7

8

9

10

11

12

Degree of Superheat ( C) Fig. 6 Variation of COP w ith Degree of Superheat

49

AU J.T. 10(1): 45-51 (Jul. 2006)

parameters. It is worth noting that the results obtained from the rig were very close to that of the model. This justifies the fact that balancing point is very important and can be used to minimize the effect of evaporator hunting. Since the characteristic curves show the same trend or pattern and performances normalized at the later stage of superheat, it means that the model itself is verified and the validity of the experimental processes confirmed. The discrepancies noted between the rig and the model values of parameters might be due to heat loss which could not be controlled since there is no perfect insulator of heat, and the defect or deficiency of the measuring instruments.

References Akintunde, M.A. 2003. Development of a vapour compression refrigeration system based on balanced points between operational units. PhD Thesis, Federal Univ. Technol., Akure, Nigeria. Akintunde, M.A. 2004. Theoretical design model for vapour compression refrigeration systems. AMSE Periodicals, France (in press). Ibrahim, G.A. 2001. Effect of sudden changes in evaporator external parameters on a refrigeration system with an evaporator controlled by a thermostatic expansion valve. Int. J. Refrig. 24: 566-76. Jabardo, J.M.S.; Mamani, W.G.; and Ianalla, M.R. 2002. Modeling and experimental evaluation of an automotive air-conditioning system with a variable capacity compressor. Int. J. Refrig. 25: 1157-72. Jung, D.; Park, C.; and Park, B. 1999. Capillary tube selection for HCFC-22 Alternatives. Int. J. Refrig. 22: 604-14. Kim, S.G.; Kim, M.S. and Ro, S.T. 2002. Experimental investigations of the performance of R22, R407 and R410A in several capillary tubes for air-conditioning. Int. J. Refrig. 25: 521-31. Lee, J.H.; Bae, S.W.; Bang, K.H. and Kim, M. H. 2002. Experimental and numerical research on condenser performance for R-22 and R-407c refrigerants. Int. J. Refrig. 25 372-82. Meyer, J.J. and Dunn, W.E. 1998. New insight into the behaviour of a metastable region operating capillary tube. Int. J. Refrig. 4 1015. Mithraratne, P.; and Wijeysundera, N.E. 2002. Dynamic simulation of a thermostaticallycontrolled counter-flow evaporator. Int. J. Refrig. 23: 174-89. Mithraratne, P., Wijeysundera, N.E., and Bong, T.Y. 2000. Simulation of evaporator stability and control. Int. J. Refrig. 22: 7489. Mithraratne, P. and Wijeysundera, N.E. 2003. An experimental and numerical study of hunting in thermostatic-expansion-valvecontrolled evaporator. Int. J. Refrig. 25: 992-8.

Conclusion The present study has examined the effect of balanced point in the oscillatory performance of evaporators. A design model based on balanced points between the operational components of vapor compression refrigeration systems (Ref-2003) developed by Akintunde (2003) was used to interpret the result obtained from the constructed rig. In addition, test data have been used to assess the quality of the computer simulation results. The simulated and the experimental performance results were compared and the results obtained from the experimental investigations justified adequately the developed design. From the experimental investigations, it could be concluded that the balancing of the operational components enhanced the performance of the evaporator. With the system balancing, the hunting of the evaporator was reduced to the barest minimum. It was noted also that, as the degree of superheat increased, the system performance was reduced. This shows that higher degree of superheat should not be allowed for optimum performance of any vapor compression refrigeration system. Present analysis and experimental investigations have shown that model results are comparable to the experimental ones. Maximum absolute deviations of the rig parameters (such as: refrigerating effect, rate of heat rejection, and coefficient of performance) from the model are within the range of 16 to 19 %. 50

AU J.T. 10(1): 45-51 (Jul. 2006)

Wei, C.Z.; Lin, Y.T.L.; Wang, C.C. and Leu, J. S. 2003. An experimental study of the performance of capillary tubes for r-407c refrigerant. ASHRAE Trans.Symp.pp.634-8. Wijaya, H. 1992. Adiabatic capillary tube test data for HF-134a. Int. J. Refrig. 11: 63-71. Wijeysundera, N.E.; Mithraratne, P., and Bong, T.Y. 2000. Simulation of evaporator stability and control. Int. J. Refrig. 23:17489. Wolf, D.A.; and Pate, M.B. 2001. Capillary tube-suction line heat exchanger performance with alternative refrigerants. ASHRAE Trans. 97: 139-49. Yasuda, H.; Machielsen, C.H.M. and Touber, S. 1983. Simulation model of a vapor compression refrigeration system. ASHRAE Trans. 89: 404-24.

Motta, S.F.Y; Parise, J.A.R.; and Braga, S.L. 2002. A visual study of r-404a/oil flow through adiabatic capillary tubes. Int. J. Refrig. 25: 586-96. Najork, H. 1997. Investigation on the dynamical behavior of evaporators with thermostatic expansion valves. Int. J. Refrig. 21: 759-69. Stoecker, W.F.1996. Stability of an evaporatorexpansion valve control loop. ASHRAE Trans.. 2: 72-85 Stoecker, W.F.; and Jones, W.P. 1982. Refrigeration and Air-Conditioning. 2nd ed. McGraw-Hill , New York,NY, USA. Wedekind, G.L. 1997. An experimental investigation into the oscillatory motion of mixture-vapor transition point in horizontal evaporator flow. J. Heat Transfer 93: 47-54.

51