Scholarly article on topic 'Effect of evaporator temperature on vapor compression refrigeration system'

Effect of evaporator temperature on vapor compression refrigeration system Academic research paper on "Mechanical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Alexandria Engineering Journal
OECD Field of science
Keywords
{"Vapor compression system" / Evaporator / Optimization / Overheat / "Binary mixture" / Refrigerant / Oil / Modelling / R600a / R410A / Propane / R32 / R134a / R22 / Comparison}

Abstract of research paper on Mechanical engineering, author of scientific article — Abdullah A.A.A. Al-Rashed

Abstract This paper presents a comparable evaluation of R600a (isobutane), R290 (propane), R134a, R22, for R410A, and R32 an optimized finned-tube evaporator, and analyzes the evaporator effect on the system coefficient of performance (COP). Results concerning the response of a refrigeration system simulation software to an increase in the amount of oil flowing with the refrigerant are presented. It is shown that there is optima of the apparent overheat value, for which either the exchanged heat or the refrigeration coefficient of performance (COP) is maximized: consequently, it is not possible to optimize both the refrigeration COP and the evaporator effect. The obtained evaporator optimization results were incorporated in a conventional analysis of the vapor compression system. For a theoretical cycle analysis without accounting for evaporator effects, the COP spread for the studied refrigerants was as high as 11.7%. For cycle simulations including evaporator effects, the COP of R290 was better than that of R22 by up to 3.5%, while the remaining refrigerants performed approximately within a 2% COP band of the R22 baseline for the two condensing temperatures considered.

Academic research paper on topic "Effect of evaporator temperature on vapor compression refrigeration system"

Alexandria Engineering Journal (2011) 50, 283-290

FACULTY OF ENGINEERING ALEXANDRIA UNIVERSITY

Alexandria University Alexandria Engineering Journal

www.elsevier.com/locate/aej www.sciencedirect.com

ORIGINAL ARTICLE

Effect of evaporator temperature on vapor compression refrigeration system

Abdullah A.A.A. Al-Rashed*

Public Authority for Applied Education and Training, Industrial Training Institute, 13092, Kuwait

Received 24 February 2010; accepted 16 August 2010 Available online 5 February 2012

KEYWORDS

Vapor compression system;

Evaporator;

Optimization;

Overheat;

Binary mixture;

Refrigerant;

Modelling;

R600a;

R410A;

Propane;

R134a;

Comparison

Abstract This paper presents a comparable evaluation of R600a (isobutane), R290 (propane), R134a, R22, for R410A, and R32 an optimized finned-tube evaporator, and analyzes the evaporator effect on the system coefficient of performance (COP). Results concerning the response of a refrigeration system simulation software to an increase in the amount of oil flowing with the refrigerant are presented. It is shown that there is optima of the apparent overheat value, for which either the exchanged heat or the refrigeration coefficient of performance (COP) is maximized: consequently, it is not possible to optimize both the refrigeration COP and the evaporator effect. The obtained evaporator optimization results were incorporated in a conventional analysis of the vapor compression system. For a theoretical cycle analysis without accounting for evaporator effects, the COP spread for the studied refrigerants was as high as 11.7%. For cycle simulations including evaporator effects, the COP of R290 was better than that of R22 by up to 3.5%, while the remaining refrigerants performed approximately within a 2% COP band of the R22 baseline for the two condensing temperatures considered.

© 2012 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V.

All rights reserved.

* Tel.: +965 97300086.

E-mail address: bdaloy@yahoo.com

1110-0168 © 2012 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. All rights reserved.

1. Introduction

Increasing concerns about climate change provide a new design factor for conventional systems striving for high efficiency and energy conservation at a given production cost. This new factor is the preference to utilize refrigerants that have a low global warming potential (GWP). Considering that the system's indirect contribution to climate change (CO2 emissions from fossil fuel power plants generating electricity to drive the system) is dominant for most applications, it is important to be able to accurately determine performance merits of different fluids, and in particular their performance potential in optimized equipment. The goal of this study was to develop optimized refrigerant designs for R600a (isobutane), R290

Peer review under responsibility of Faculty of Engineering, Alexandria University.

doi:10.1016/j.aej.2010.08.003

Nomenclature

COP coefficient of performance x vapor quality

G refrigerant mass flux (kg s-1 m-2) l viscosity (iPa s)

GWP global warming potential P density (kg m~3)

h enthalpy (kJ kg-1) g efficiency

hfg latent heat (kJ kg-1)

k thermal conductivity (W m-1 K-1) Subscripts

m mass flow rale (kg s-1) dis discharge

mr refrigerant mass flow rate (kg h-1) e-m electro-mechanical

P pressure (kPa) f saturated liquid

Q total capacity (kW) g saturated vapor

Qi latent capacity, portion of total capacity due water in evaporator inlet

vapor removal (kW) is isentropic

s* normalized entropy m refrigerant-oil mixture

Tsat saturated temperature at the evaporator exit (0C) mot motor

V volumetric flow rate (m3/s) out evaporator outlet

W power (W) suct suction

(propane), R134a, R22, R410A, and R32 finned-tube evaporators and to analyze the effect of optimization on evaporator and system performance for these refrigerants. The number of studies related to the effects of the lubricating oil in vapor compression refrigeration systems has been increased in last few years. The renewed interested for this subject is linked to the replacement of the usual couples of hydrochlorofluorocar-bon (HCFC) and mineral oils by hydrofluorocarbon (HFC) and synthetic lubricants (either polyolester oil (POE), polyalk-yleneglycol oil (PAG) or even polyvinylester (PVE)). The lubricants used with vapor compression refrigeration systems prevent the wear of the compressor, limit the heating of the refrigerant during compression and take part in the sealing of the whole of the circuit. Although essential to the correct operation of the system, the use of lubricants is accompanied by adverse effects, which depend, amongst other, on the chemical compatibility between oil and refrigerant and on the rate of release of this oil out of the compressor. These effects are linked to modification of the physical and thermodynamic properties of the refrigerant-oil mixture, which can have a significant effect on the quality of heat transfer within the heat exchangers or on the characteristics of the flows. A first part of the work, presented recently [1-3] led in the conception of a refrigeration system simulation software that takes into account the presence of the oil rejected by the compressor. The system whose operation is simulated uses the R410A HFC blend and its compressor is lubricated by an ISO 32 POE synthetic oil. It is a fully instrumented laboratory prototype, which provides a lot of experimental results that can be compared to computed values. The numerical and experimental results associated to seven working points were analysed in detail and it appeared that there is no significant difference between them: the consequences of the presence of lubricant are quantitatively identical, whatever the pressure ratio, the evaporating and condensing temperatures. However, the control of the temperature increase in the evaporator seemed to influence greatly the performance of the system, when no oil is circulating, which was expected, but also when some oil is circulating. It must be noted that when some lubricant circulates in the system, the difference between the temperatures at the evaporator

inlet and outlet represents only an apparent gas overheat, since the quality never equals 1. In the absence of oil, the gas overheat must of course be reduced to a minimum (provided it stays greater than the glide of the R410A) but in the presence of oil, optimal values of the apparent overheat are expected: indeed, simple enthalpy models [4] validated by experiments [5] show that, for fixed oil fraction and mass flow rate, the enthalpy change in the evaporator is an asymptotically growing function of the temperature increase; however, in real systems operating with constant temperature heat source and evaporator of constant exchange surface, an excessive overheat reduces the refrigerant mass flow rate, the evaporator effect and the refrigeration coefficient of performance (COP).

2. Refrigerant studied

Table 1 presents the studied refrigerants in the order of their saturation vapor pressure corresponding to 7.0 0C dew-point temperature. The selected refrigerants have different properties. The liquid conductivity and viscosity of the studied refrigerants, the most influential properties for refrigerant's heat transfer and pressure drop, differ by as much as 15% and 110%, respectively. Greater differences, however, are seen in the thermodynamic properties: the vapor densities differ by up to a factor of 7, dTsat/dP differ by as much as a factor of 4.6, and the latent heats differ by as much as 80%. These properties are related to refrigerant's critical temperature and the shape of the two-phase dome. They affect the selection of the optimal refrigerant mass flux in the refrigerant circuitry and, as we will present it in the later section, refrigerant's COP in the vapor compression cycle. Fig. 1 shows a temperature-entropy diagram using a normalized entropy scale to facilitate qualitative comparison of impact of thermodynamic properties on the COP for the studied refrigerants. Fig. 2 shows that the Muller-Steinhagen and Heck correlation agrees very well with the modified Pierre correlation. Compared to the Muller-Steinhagen and Heck correlation, the Pierre correlation has the disadvantage that it is not applicable to adiabatic flows. Also, the Pierre correlation calculates the overall pressure drop in a heat

Table 1 Refrigerant informations [5].

Refrigerant Pg (kPa) Pf(kgm 1) Pg(kgm 1) hfg (kJ kg ') kf (W m -1 K-1) If (iPa s) dTsat/dP (KkPa-1) GWP (100 years)

R600a 199.5 5.34 572.2 348.2 0.0958 183.05 0.1477 20

R134a 374.6 18.32 1271.3 193.2 0.0889 243.88 0.0770 1320

R290 584.4 12.69 519.0 364.5 0.1024 116.89 0.0585 20

R22 621.5 26.35 1257.3 199.3 0.0916 200.13 0.0516 1780

R410A 995.0 38.19 1141.7 212.6 0.1056 154.92 0.0329 2000

R32 1011.5 27.56 1030.6 304.0 0.1398 139.24 0.0322 543

Thermo-physical properties are for saturation temperature of 7.0 °C; based on Ref. [5].

Figure 1 Temperature-entropy diagram for studied refrigerants (entropy is normalized to the width of the two-phase dome, i.e. sg = 0 and sg = 1).

Figure 2 Comparison of nine pressure drop relations.

exchanger and cannot predict local pressure drop values, especially at high quality range approaching the saturated line.

3. Evaporator performance with selected refrigerants

Table 2 gives the evaporator design data that was common for all evaporator simulations in this study. Additionally, the air

Table 2 Evaporator design information.

Items Unit Value

Tube length mm 500

Tube inside diameter mm 9.2

Tube outside diameter mm 10.0

Tube spacing mm 25.4

Tube row spacing mm 22.2

Number of tubes per row 12

Number of depth rows 3

Fin thickness mm 0.2

Fin spacing mm 2

Tube inner surface Smooth

Fin geometry Louver

Air volumetric flow rate 3 • — 1 m3 min 1 25.5

condition was 26.7 0C dry-bulb temperature and 50% relative humidity. The refrigerant inlet condition was specified in terms of the saturation temperature and sub-cooling at the inlet to the distributor, which was included in the simulation runs. We used sub-cooling of 5.0 K in all simulations. With specified inlet parameters and environmental conditions, evaporator iterated refrigerant mass flow rate to arrive with a 5.0 K refrigerant exit superheat for the specified exit saturation temperature.

The first simulation task was to obtain evaporator capacity for each refrigerant at the same exit saturation temperature of 7.0 0C. Because of significant differences in thermo physical properties, refrigerant circuitry had to be optimized for each refrigerant. We started by manually developing five basic circuitry architectures involving 1, 1.5, 2, 3, and 4-circuits, four of which are shown in Figs. 3 and 4 presents capacity results for the prearranged 1, 1.5, 2, 3, 4-circuit designs and the optimized designs developed by ISHED1 (intelligent system for heat exchanger design). For each refrigerant, the design developed by ISHED1 which is the best of the prearranged designs. For R32, R410A, R290, and R22, ISHED1 developed individually optimized designs, which were based on a 1.5-circuit. Although each of these designs had a somewhat different layout, EVAP (evaporator simulation model EVAP from the EVAP-COND simulation package of NIST) simulations confirmed that they were equivalent in performance. For this reason, only the R410A 1.5 circuitry ISHED1-developed design was used further for R32, R410A, R290, and R22. For R134a and R600a, a 3-circuit and a 4-circuit design, respectively, were proposed by ISHED1.

EVAP simulations using ISHED1 optimized evaporators generated the results presented in Table 3. For comparative eval-

Figure 3 Manually developed 1.5, 2, 3, and 4-circuit designs (side view; circles denote tubes; continuous lines indicate return bends on the near side of the heat exchanger, broken lines indicate return bends on the far side, full circles indicate outlet tubes).

Figure 4 Evaporator capacities for manually developed and ISHED1-optimized circuitry designs.

uation, we selected R22 as our reference. R600a had the lowest capacity, 9.5% below that of R22, and R32 had the highest

capacity exceeding that of R22 by 14.5%. We also should note that the low-pressure refrigerants, R600a and R134a, had the lowest ratio of the latent capacity to total capacity.

4. Effect of evaporator performance on system COP

The basic thermodynamic analysis of the vapor compression cycle has been analyzed, as implemented by the CYCLE_D model (basic thermodynamic analysis of the vapor compression cycle, as implemented by the CYCLE_D model) [6], to assess the impact that the evaporator performance has on the COP for different refrigerants. In the CYCLE_D simulations, refrigerant saturation temperatures in the evaporator and condenser are specified as input. To acquire all of the data performed two rounds of simulations. In the first round, the same evaporator exit saturation temperature, Tsat, of 7.0 0C for each refrigerant. For the second round, we first performed iterative EVAP simulations at various evaporator saturation temperatures to obtain a capacity equal to that of R22 at 7.0 0C saturation temperature. The obtained saturation temperatures, constituted a new input for each refrigerant (instead of 7.0 0C) for the second round of CYCLE_D simulations. Perform simulations at two condensing temperatures of 38.0 and 45.0 0C. Table 4 contains the additional CYCLE_D input used in these calculations, and Table 5 presents the obtained results for the two rounds of simulations. The results in the left-hand-side of the table, with Tsat = 7.0 0C, are from the basic thermodynamic calculations of the cycle. The results located in the right-hand-side of the table, with different values of Tsat, account for the impact that the thermodynamic and transport properties have on the cycle through their effect on the performance of the optimized evaporator.

Fig. 5 presents COP results referenced to the COP of the baseline R22 cycle. As expected, COPs for theoretical simulations in the basic cycle ranked the refrigerants in the order of their critical temperatures (Fig. 1). For the condensing temperature of 38 0C, COP of R600a is 5.3% better than that of R22, and the COP of R32 is 5.1% worse. However, the performance of the group was found to be much more uniform when the effects of the optimized evaporators and corresponding saturation temperatures are included in the simulations. While propane arrived as the most efficient refrigerant with a 3.5% better COP than R22, the COPs of the remaining refrigerants were found to be within 0.7% of the COP of R22. The high-pressure (low critical temperature) R32 experienced the greatest COP improvement when evaporator effects were taken into account, and in relation to R600a it changed the 10.4% COP deficit to 0.8% advantage. All refrigerants provided similar latent capacities. The results for the 45.0 0C condensing temperature display similar trends with the difference that the high-pressure refrigerants (R32 and R410A) showed somewhat lower performance because the cycle moved closer to their critical points.

5. System modeling

The software on which the optimum searches are performed was developed to simulate the operation of an experimental prototype, which architecture is usual. A thermostatic expansion valve regulates the refrigerant flow into the evaporator and a bottle of liquid is inserted between the condenser and a small subcooler. The compressor is of the piston type and

Table 3 Summary of simulation results for ISHEDl-optimized designs for Tsat = 7.0 °C [8].

Refrigerant Number of circuits xin (-) Pout (kPa) DP (kPa) DTsat (K) m, (kg h ) Q (kW) Q/QR22

R600a 4 0.26 200 12 1.7 102.0 7.430 0.905

R134a 3 0.27 375 27 2.0 195.6 7.787 0.948

R290 1.5 0.27 585 59 2.8 116.1 8.706 1.060

R22 1.5 0.23 621 64 3.2 190.7 8.211 1.000

R410A 1.5 0.29 993 57 1.8 213.5 9.091 1.107

R32 1.5 0.24 1012 40 1.3 143.0 9.399 1.145

Table 4 Input data to CYCLE_D.

Inputs Unit Data

Compressor isentropic efficiency 0.65

Type of the compressor is hermetic

Compressor volumetric efficiency 0.82

Electric motor efficiency 0.85

Suction line pressure drop °C 1.0

Discharge line pressure drop °C 1.0

Evaporator superheat °C 5.0

Condenser sub-cooling °C 5.0

Liquid line-suction line heat exchanger None

has a swept volume of 13 m3/h. The evaporator and the condenser are plate heat exchangers. The external heat carrier fluid is water or a mixture of water and a cryoprotectant (mono propylene glycol).

The numerical model simulates the operation of the prototype in the stationary mode. The amount of lubricant flowing in the circuit can vary between any negligible value and 5% (in weight of refrigerant). The external parameters that define the operation of the system are the mass flow rate and the inlet temperatures of water in the exchangers.

The operation of the compressor is characterized by three parameters: the volumetric efficiency gvol (1), the isentropic efficiency gis (2), and the electro-mechanical efficiency ge-m (3), which are correlated as function of the pressure ratio by analysis of experimental data (in the absence of oil)

hdis,r hsuct,r

m(hdis,m - hSuct,m) " Wmot

gvol :

The oil coming from the evaporator flows in liquid state in the compressor and assuming that this do not change the definition and the value of the volumetric efficiency gvol and of the isentropic efficiency gis [7-9].

Further details about the compressor modelling, the heat exchangers modelling and the numerical results validation by comparison to experimental data can be found in Refs. [10-12].

6. Results

All the results presented below were obtained for water mass flow rates of 1.4 and 1.0 kg/s, in the condenser and evaporator, respectively. The inlet water temperature in the evaporator is 17.3 °C, while it is 43.4 °C in the condenser. The modification of these parameters would not change the comments and conclusions. Figs. 6-8 show the influence of the apparent overheat on the heat exchanged at the evaporator, on the power supplied to the compressor, and on the refrigeration COP. The apparent overheat varies from 1.5 to 12 K and different curves are plotted, depending on the amount of lubricant circulating, which ranges from 0 to 5%. As expected, the performances of the system decreases with an increasing amount of lubricant.

It can be seen in Fig. 6 that for a given amount of circulating oil, there is an optimal value of the apparent overheat for which the heat exchanged at the evaporator is maximized.

dis,is,r

suct,r

Table 5 Performance for the theoretical cycle and the cycle accounting for evaporator effects.

Refriger ant Basic theoretical cycle Cycle including evaporator effects

Tsat (°C) COP Tsat (°C) COP Qfactor

38.0 °C Condensing temperature

R600a 7.0 4.103 5.7 3.895 0.22

R134a 7.0 3.993 6.4 3.896 0.22

R290 7.0 3.929 7.7 4.036 0.21

R22 7.0 3.898 7.0 3.898 0.21

R410A 7.0 3.703 8.1 3.874 0.21

R32 7.0 3.701 8.5 3.926 0.21

45.0 °C Condensing temperature

R600a 7.0 3.237 5.8 3.111 0.22

R134a 7.0 3.133 6.4 3.064 0.22

R290 7.0 3.074 7.8 3.155 0.21

R22 7.0 3.063 7.0 3.063 0.21

R410A 7.0 2.869 8.2 2.995 0.21

R32 7.0 2.878 8.5 3.073 0.21

Figure 5 COPs compared to the COP of R22 for the basic cycle and for the cycle including evaporator effects for 38.0 and 45.0 °C condensing temperatures.

Figure 6 Changes in the cooling capacity when the gas apparent overheats are varying for different mass fraction of oil polluting the circuit.

However, in the presence of oil, this optimum is located at values of the overheat that vary between 1.8 K {w = 0.01%) and 3.3 K {w = 5%), which is less than what the expansion valves usually impose in dry-expansion evaporators (Fig. 9), especially at low oil concentrations, the slope of the curves around the optimum is very small, meaning that when the apparent overheat is below 6 K, the performances of this particular system are limited rather by the other components than by the evaporator. Observing the curves of Fig. 6, it can be concluded that an apparent overheat below 6 K in the evaporator would not, in the presence of oil, improve significantly the exchanged energy.

Figure 7 Changes in the compressor power when the gas apparent overheat is varying.

DC Ï.1

Refrigeration COP

.......——

^................

JSS fraction - 5.0 % iss fraction _ 2.5 %

■ Oil m

* Oil mass fraction - 1,0 % Apparent

- Oil mass fruition - 0.00 % overheat (°C)

1.5 3 4.5 « 7.5 9 10.5 12

Apparent Overheat Temperature (C)

Figure 8 Refrigeration COP as a function of the gas apparent overheat in the evaporator for various oil mass fraction in the circuit.

Figure 9 Apparent gas overheat in the evaporator for optima of the refrigeration COP and evaporator energy as functions of the mass fraction of lubricant polluting the circuit.

Fig. 7 shows that the compressor power diminishes with an increasing apparent overheat, which is a direct consequence of the refrigerant mass flow rate reduction induced by the closing of the expansion valve. It can also be seen in Fig. 7 that the

power needed by the compressor increases with a rising amount of circulating oil. Two reasons can explain these facts: at first, the heat capacity of the lubricant is high, and the energetic cost associated with its heating in the compressor is higher than the compression of the refrigerant. For instance, when the pressure varies from 1.0 to 3.0 MPa and when the temperature increases from 15 to 50 0C, too high specific enthalpy of R410A is 52 kJ/kg while it reaches about 142 kJ/kg for the liquid oil. However, this result depends strongly on the hypotheses used for the modeling of the compressor and, in particular, on the calculation of the enthalpy of the refrigerant-oil mixture at the discharge port. In our case, we suppose that the temperature of the mixture is only a function of the gas properties at the suction port and of the pressure ratio. The second reason is linked to the solubility of the refrigerant in liquid oil, which decreases when the pressure and temperature increase. This means that some of the refrigerant will evaporate in the compressor; all the more so that the apparent overheat is low. Fig. 10 shows for instance how the quality at the evaporator exit depends on the apparent overheat; on the other hand, the oil mass fraction in the remaining liquid at the compressor discharge is always close to 0.8, which means that the quality is equal to 0.9937, 0.9875, 0.9687 and 0.9875 when the total amount of lubricant circulating in the system equals 0.5%, 1%, 2.5% or 5.0%. The energetic cost associated with the quality increases in the compressor is high, due of course to the latent heat associated with this process. However, we suppose that this does not influence the discharge temperature, which is calculated as if the quality would not be the same at both the suction and discharge ports, and this hypothesis maximizes the compressor work.

An optimum can be observed in the evolutions of the refrigeration COP (Fig. 8), but due to the simultaneous decrease in the compressor and evaporator energy with the rise of the apparent overheat, this optimum is located at much higher values than the optimum of the evaporator energy. Fig. 9 clearly shows that the optima of the refrigeration COP and of the energy exchanged at the evaporator are very close when the amount of lubricant in the circuit is low (below 0.5%) but that they diverge when the quantity of oil rises. Considering these curves, it could be considered useful to modify the overheat control in order to optimize either the evaporator energy or the refrigeration COP. However, Fig. 10 suggests to keep the

Figure 10 Quality at the evaporator exit for different mass fraction of oil in the circuit and varying overheat.

Figure 11 Loss of evaporator energy and COP when the gas apparent overheat is adjusted to the other optimum.

apparent overheat as high as possible, in order to maximize the quality at the evaporator exit and to prevent the risks of compressor damage. Another argument in the favor of a higher overheat comes from Fig. 11, where one can see that the loss of the evaporator energy is less important than the loss of the refrigeration COP when the apparent overheat is adjusted to the other optimum.

7. Conclusions

The results of a refrigeration system simulation software to an increase in the amount of oil flowing with the refrigerant show that there are optimal values of the apparent overheat, for which either the exchanged heat or the refrigeration COP is maximized. It is not possible to optimize both the refrigeration COP and the evaporator energy. However, in most of the cases the optima values of the apparent overheat are below the values that the expansion valves usually impose. It would be possible to improve the refrigeration COP by increasing the apparent overheat above 6 K when the oil mass fraction is greater than 1%, but at the expense of the evaporator energy. In this study, we evaluated the performance of R600a, R134a, R290, R22, R410A, and R32, which differ vastly in critical temperatures and other thermo physical properties. We optimized evaporator circuitry for each refrigerant using a non-Darwinian evolutionary scheme, and performed simulations of the optimized evaporators. The high-pressure refrigerants provided higher evaporator capacities than the low-pressure refrigerants. For a 7.0 0C evaporator exit saturation temperature, and using R22 as a reference, R32, R410A, and R290, had a greater capacity by 14.5%, 10.7%, and 6.0%, while R134a and R690a had a lower capacity by 5.2% and 9.5%, respectively. The subsequent theoretical cycle simulations with the same 7.0 0C evaporator saturation temperature showed the COPs of the studied refrigerants to be in the order of their critical temperatures, i.e. low-pressure refrigerants had the best COPs. However, for the cycle simulations including evaporator effects (carried out at a different evaporator saturation temperature for each fluid to match the R22 capacity), the refrigerants performed within approximately a 2% band of the R22 COP baseline for the two condensing temperatures used. The exception to this was R290, whose COP was better than that of R22 by approximately 3% due to a set of favorable thermo physical properties. It is worth noticing that R32 overcame the 10% COP deficit it had in the theoretical cycle in reference to

R600a and showed a comparable performance when evaporator effects were included in the cycle simulation. It must be emphasized that this study isolated the evaporator effects, and did not include similar effects that may be introduced by the condenser. Also, we have to note that selection of the compressor and relative sizing of the remaining components will affect the performance of a complete system. This study was not concerned with design and the cost related to the practical implementation of different refrigerants, equipment size, pressure, or lubricant issues. The condensing and evaporating temperatures used in this study correspond to the comfort cooling application. An additional insight could be obtained from a similar study performed at the same reduced temperatures for the considered refrigerants.

References

[1] O. Lottin, P. Guillemet, J.M. Lebreton, Effects of synthetic oil in a compression refrigeration system using R410A - Part I: Modelling of the whole system and analysis of its response to an increase in the amount of circulating oil, International Journal of Refrigeration 26 (7) (2003) 772-782.

[2] O. Lottin, P. Guillemet, J.M. Lebreton, Effects of synthetic oil in a compression refrigeration system using R410A - Part II: Quality of heat transfer and pressure losses within the heat exchangers, International Journal of Refrigeration 26 (7) (2003) 783-794.

[3] O. Lottin, P. Guillemet, J.M. Lebreton, M. Feidt, Influence of oil on the operation of vapour-compression refrigeration systems, in: 21st Int Cong Refrigeration, Washington, August 17-22, 2003.

[4] M. Youbi-Idrissi, J. Bonjour, C. Marvillet, F. Meunier, Impact of refrigerant-oil solubility on an evaporator performances working with R-407C, International Journal of Refrigeration 26 (3) (2003) 284-292.

[5] M. Youbi-Idrissi, J. Bonjour, C. Marvillet, M.F. Terrier, F. Meunier, Oil presence in an evaporator: experimental validation of a refrigerant/oil mixture calculation model, International Journal of Refrigeration 27 (3) (2004) 215-224.

[6] P.A. Domanski, D.A. Didion, J. Chi, NIST Vapor Compression Cycle Design Program - CYCLE_D 3.0, Standard Reference Database 49, National Institute of Standards and Technology, Gaithersburg, MD, 2003.

[7] R.S. Michalski, Learnable evolution model: evolutionary process guided by machine learning, Machine Learning 38 (1) (2000) 9-40.

[8] M.B. Ould Didi, N. Kattan, J.R. Thome, Prediction of two-phase pressure gradients of refrigerants in horizontal tubes, International Journal of Refrigeration 25 (7) (2002) 935-947.

[9] NIST, Simulation Models for Finned-Tube Heat Exchangers -EVAP-COND, National Institute of Standards and Technology, Gaithersburg, MD, 2003.

[10] P.A. Domanski, D. Yashar, K.A. Kaufman, R.S. Michalski, Optimized design of finned-tube evaporators using learnable evolution methods, International Journal of HVAC&R Research 10 (2) (2004) 201-212.

[11] ASHRAE, ANSI/ASHRAE Standard 34-2001, Designation and Safety Classification of Refrigerants. American Society of heating, Refrigerating, and Air-Conditioning Engineers, Atlanta, GA, 2001.

[12] ASHRAE, ANSI/ASHRAE Fundamentals, American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Atlanta, GA, 2005.