Energy and Buildings 36 (2004) 690–695

Combined air conditioning and tap water heating plant using CO2 as refrigerant Willy Adriansyah Thermodynamic Research Laboratory, Inter University Research Center, Institut Teknologi Bandung, 40132 Bandung, West Java, Indonesia

Abstract A combined air conditioning and tap water heating plant using carbon dioxide (CO2 ) as refrigerant has been investigated theoretically and experimentally. The system is suitable for countries with year around cooling demand, such as Indonesia or Singapore, and a need for hot tap water. A unique CO2 transcritical cycle characteristic for heating process can afford an improvement to a CO2 air conditioning system when rejected heat from the system is recovered. Some parameters affecting performance of the combined system are discussed. © 2004 Elsevier B.V. All rights reserved. Keywords: Combined ac and water heating; Carbon dioxide; Transcritical cycle; Optimum pressure

1. Introduction The use of natural refrigerant after CFC era has been attracting many research institutions and related industries. Among common substances that can be used as refrigerant, such as air, NH3 , carbon dioxide (CO2 ) or R-744, hydro carbon, and water, CO2 has unique characteristics and almost fulfill all required properties to be used as refrigerant. It has zero ODP and negligible GWP, high heat transfer coefficient, excellent availability, compatible with material of refrigeration system, very low cost, and—the most important—free from refrigerant market monopoly [1]. However, there is one short of drawback. The operating pressure is high, around eight times compared to that of conventional refrigerants. This needs completely new design of system components and with current construction and production technology the components can be produced [2]. Various applications of CO2 as refrigerant in transcritical cycle have been investigated and some of them show promising result, especially for heat pump [3]. For air conditioning application, the performance of CO2 is comparable to that of R-22 or R-134a [4]. The other interesting potential of CO2 is for producing simultaneous cooling and heating. Such system could yield superior total system efficiency (ratio of cooling load and heating load to compressor power consumption) because of the nature of transcritical cycle which absorbs energy at constant temperature and rejects energy at gliding temperature. This type of applica-

E-mail address: [email protected] (W. Adriansyah). 0378-7788/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2004.01.014

tion is suitable in tropical countries such as Indonesia and Singapore, where cooling is needed year around. In a building which needs air conditioning and water heating, use of rejecting heat from an air conditioning system to produce hot water could conserve energy. Depending on the type of buildings and occupant culture, recovered energy from the air conditioning system could be quite high. In this paper, a combined air conditioning and tap water heating system operates in transcritical cycle will be discussed.

2. Transcritical cycle The word transcritical comes from the characteristic of this cycle. Heat is absorbed at constant temperature below the critical temperature while heat rejection process takes place at pressure above the critical pressure. The reason why CO2 should be operated in this cycle is that the critical temperature of CO2 is very low, i.e. 31 ◦ C and for most air conditioning system run in hot climate, most of outdoor air temperature is around or above this value. Even when the air conditioning system could be run below the critical point, refrigerating capacity will be very low. Fig. 1 shows transcritical cycle on Ph diagram on which effect of heat rejection pressure is depicted. The corresponding flow sheet is shown to the right on the figure. Since heat is rejected at supercritical region, pressure and temperature is independence and for a specified temperature the pressure can be changed. There is no condensation process in this cycle and heat rejecting device is named gas cooler. As can be seen on Fig. 1, for the same outlet

W. Adriansyah / Energy and Buildings 36 (2004) 690–695

691

Fig. 1. Transcritical cycle on Ph diagram and the corresponding flow sheet.

temperature of the gas cooler (point 3 or 3 ), compressor discharge pressure can be changed from pressure at point 2 to pressure at point 2 . It can also be observed that specific refrigerating capacity will change as discharge pressure changes. It means the refrigerating capacity can be controlled by changing discharge pressure. This capacity control cannot be applied in a conventional refrigeration cycle because all processes occur below critical point. Since altering discharge pressure will change both specific refrigerating capacity and compressor power, coefficient of performance (COP) will vary with the discharge pressure. Fig. 2 shows variation of COP with discharge pressure, where ‘qe’ means specific refrigerating capacity, ‘wc’ means specific compressor power, and ‘cop’ means coefficient of performance. Fig. 2 also shows that specific refrigerating capacity increases as discharge pressure increases. The increase in specific refrigerating capacity is more than the increase in specific compressor power consumption until it reaches a point where the increase is slower. Meanwhile, compressor power consumption increases more or less linear with discharge pressure. Hence, COP first increases until a point and then decreases. The pressure where COP reaches maximum value is called optimum pressure. Under normal operation, transcritical cycle should be run at the optimum

Fig. 2. Performance variations at various discharge pressures.

pressure. When more cooling effect is needed for short time such as at pulling down period, discharge pressure could be increased beyond the optimum pressure by sacrificing COP.

3. Combined air conditioning and tap water heating Transcritical cycle is ideal for heating process. Rejecting heat at supercritical region means heat transfer occurs by sensible cooling in the gas cooler. Temperature of CO2 will decrease continuously from discharge temperature to temperature out of the gas cooler. Because there is no condensation, high hot water temperature can be achieved. Hot water up to 90 ◦ C can be produce easily while heating-COP still high [3]. Such high hot water temperature cannot be achieved by a conventional heat pump system without running the system at very low heating-COP. Fig. 3 shows two heat rejection processes in a transcritical cycle using CO2 and in a conventional refrigeration cycle using R-22. Note that pinch temperature (the smallest temperature difference in a heat exchanger) occurs somewhere inside heat exchanger in conventional cycle due to condensation process, while it can occur at the cold end of heat exchanger in transcritical cycle. Temperature different at the cold end of heat rejection process is very important factor which affects COP. This temperature different is named temperature approach. For the same cooling medium temperature, the lower temperature approach the higher the COP will be. On Fig. 3, it can be seen that the temperature approach is lower in CO2 system than in R-22 system because pinch temperature occurs at the cold end of heat exchanger. From experimental evidence it is found that CO2 has excellent performance in a hot water heat pump system while rather inferior in an air conditioning system compared to conventional cycle, transcritical cycle operated as a combined air conditioning and heat pump would be an interesting application in areas where there is a need for cooling and heating simultaneously. This combined system offers both saving in energy consumption for producing hot water

692

W. Adriansyah / Energy and Buildings 36 (2004) 690–695

Fig. 3. Heat rejection process in trancritical cycle (left) and conventional cycle (right).

Fig. 4. Ideal combined system with one gas coolers.

and improvement of the performance of the air conditioning system. Fig. 4 shows a schematic of an ideal combined system. All rejected heat is utilized for water heating process and of course it will have the highest efficiency. In application where not all rejected heat is captured, and in fact this is most of the case, an additional gas cooler is needed to reject the rest of heat as shown in Figs. 5 and 6. There can be two possible arrangements of these gas coolers, series (Fig. 5) and parallel (Fig. 6). The parallel arrangement gives flexibility in controlling capacity of the gas cooler for producing hot water. The capacity can be controlled by regulating distribution of CO2 flowing to both gas coolers. Another advantage of the parallel configuration is a higher improvement of the air conditioning performance. CO2 temperature leaving the gas coolers will become lower as percentage of heat recovery increases at condition where inlet water temperature is lower than cooling medium temperature. As this temperature becomes lower, the air conditioning performance becomes higher.

On the other hand, the capacity is fixed for operating condition in the series arrangement. Cooling-COP will be dictated by cooling medium temperature and enhancement of the air conditioning side is negligible. Despite a minor enhancement, there is an advantage of the series arrangement.

Fig. 5. Combined system with series configuration.

W. Adriansyah / Energy and Buildings 36 (2004) 690–695

693

Table 1 Relative uncertainties for two different discharge pressures

Fig. 6. Combined system with parallel configuration.

Heat transfer area of the additional gas cooler for the same hot water load is smaller compared to the parallel arrangement. This is because the entire mass of CO2 from the compressor flows through the additional gas cooler.

4. Prototype of combined air conditioning and water heating system Test rig of the combined air conditioning and tap water heating system is a modified version of hot water heat pump system developed by Zakeri et al. [5] with an addition of a water-cooled gas cooler and its water loop. The original version of the test rig was designed for heat pump water heater using CO2 with heating capacity of 50 kW at 0 ◦ C evaporation temperature, 7 ◦ C inlet water temperature, and 60 ◦ C hot water temperature. The new water loop together with a water-cooled gas cooler is designed and built which is intended to simulate different heat sink conditions. The two water-cooled gas coolers in the test rig can be arranged in series or parallel by switching CO2 flows in the pipeline in order to observe the characteristic of the system in different gas coolers configuration. Fig. 7 shows photograph of the test rig.

Parameter

Ph = 85 bar (%)

Ph = 110 bar (%)

Compressor power Air-cooled gas cooler capacity Evaporator capacity Refrigerant flow rates Cooling-COP Isentropic efficiency Volumetric efficiency

±0.17 ±4.16 ±4.66 ±4.35 ±4.66 ±4.69 ±4.37

±0.15 ±3.73 ±3.88 ±3.79 ±3.88 ±4.07 ±3.81

5. Instrumentation and measurement accuracy All temperature measurements are carried out by type-T thermocouple with accuracy ±0.5 ◦ C. All thermocouples are connected to a data logger that converts voltage signal from the thermocouples to temperature. DRUCK pressure transducers are used to measure absolute pressures, which send voltage signal to the data logger. To obtain more accurate measurement, differential pressure transducers are used to measure pressure drop across the heat exchangers. Accuracy of these absolute transducers is ±0.1% of measured value and accuracy of the differential pressures is 0.04% of measured value. Water flows through the heat exchangers are measured by turbine flow meters. The accuracy of the flow meters are 0.2% of calibration span times water density. CO2 flow rates are determined through heat balance calculation in both gas coolers. Three-point thermopile was installed to measure temperature different across the gas coolers so that its accuracy becomes ±0.3 ◦ C. However, in the final report all uncertainties calculation is based on ±0.5 ◦ C temperature accuracy. A rotation meter and a torque meter are installed through which the compressor power consumption is calculated. Accuracy of the rotation meter is ±1 rpm and accuracy of the torque meter is ±0.5% of measured value. Uncertainties of the experimental data are depending on the discharge pressure, especially when running at a pressure around the critical region of CO2 at which the uncertainties become higher. Most of the optimum conditions are in a range from 85 to 105 bar discharge pressure and the relative uncertainties for cooling capacity, gas cooler capacity and cooling-COP are around 5%. Table 1 shows the uncertainties of the experimental results for three discharge pressures at 0 ◦ C evaporation temperature and 30 ◦ C cooling medium temperature.

6. Method of controlling load ratio

Fig. 7. Prototype of the combined air conditioning and water-heating system.

For parallel configuration, load ratio or percentage of heat recovery, ‘xr’ is set at desired value. For a discharge pressure, ˙ w ’ can be calculated from the following hot water load, ‘Q relationship:

694

W. Adriansyah / Energy and Buildings 36 (2004) 690–695

˙0 ˙ w = xr Q Q

ac bmode

˙ 0 ac mode ’ is total rejected heat of air conditionwhere ‘Q ing without heat recovery at the same discharge pressure. Hence, for a certain inlet water temperature and hot water temperature, the required water mass flow rates, ‘m ˙ w ’ can be calculated through: m ˙w =

˙w Q Cpw Tw

where Cpw is specific heat capacity and Tw is temperature difference. By controlling mass flow rates of CO2 entering the additional gas cooler, hot water temperature can be set to a desired value. 7. Experimental results and discussion Fig. 8 shows cooling-COP at various load ratios. It can be seen from the figure that at all load ratios, the cooling-COP trend is the same. The main different is the location of the

optimum pressure. An advantage of parallel configuration is that the cooling-COP can be increased by increasing load ratio. The optimum cooling-COP was increased by 5.7% at 25% heat recovery up to 16.7% at full recovery mode. Owing to a large capacity of the additional gas cooler, the test rig is not suitable to be run as a combined system with series configuration. To investigate the effect of gas coolers configuration, a verified simulation program are developed and used for the series configuration. The effect of the configuration on cooling-COP is shown in Fig. 9. As seen in this figure, the increase on maximum cooling-COP of the series configuration is negligible compared to the system without heat recovery. However, in the parallel configuration, the maximum cooling-COP increase from 3.00 to 3.24 at load ratio of 0.25 (25% heat recovery) and the optimum pressure shifted to a lower value from 90 to 86.2 bar. The change in cooling capacity for the series configuration is also negligible. At the optimum pressure, the cooling capacity of the series configuration was 23.6 kW as compared to cooling capacity of the system without heat recovery of

Fig. 8. Cooling-COP variation at various load ratio [Tevap = 0 ◦ C, Tsink = 30 ◦ C, Tw

in

= 20 ◦ C, Tw

Fig. 9. Cooling-COP of the ac only, series, and parallel configuration [Tevap = 0 ◦ C, Tsink = 30 ◦ C, Tw

in

out

= 60 ◦ C].

= 20 ◦ C, Tw

out

= 60 ◦ C].

W. Adriansyah / Energy and Buildings 36 (2004) 690–695

Fig. 10. Total-COP at various heat recovery ratios [Tevap = 0 ◦ C, Tair = 30 ◦ C, Tw

23.3 kW, while the cooling capacity of the parallel configuration at optimum discharge pressure and load ratio of 0.25 was 24.6 kW. In term of overall performance for a combined air conditioning and tap water heating system, the utilization of rejected heat of the air conditioning system should be taken into account when calculating total coefficient of performance (total-COP). There will be two boundaries for the total-COP, the lower boundary when there is no heat recovery and the upper boundary when all rejected heat is utilized. Between these boundaries, the total-COP will vary depend on the percentage of heat recovery. Fig. 10 shows total-COP as a function of discharge pressure at various heat recovery ratios. As can be expected, the total-COP increases as heat recovery ratio increases. It can be seen that the optimum discharge pressure are almost the same as that of the air conditioning without heat recovery for all load ratios except for full recovery mode where the optimum pressure is higher.

in

695

= 20 ◦ C, Tw

out

= 60 ◦ C].

consumption of hot water system is quite high in high rise building, the combined system offers promising potential as an energy recovery system where energy consumption of the water-heating system can be reduced significantly or be eliminated completely when all rejected heat is utilized for producing hot water.

Acknowledgements This paper is part of doctoral work carried out at Department of Refrigeration and Air Conditioning, Norwegian University of Science and Technology (NTNU). The study was funded for the first semester by Direktorat Jenderal Pendidikan Tinggi Indonesia and the rest by Ministry of Education, Research, and Church Affairs (KUF) in cooperation with International Office of NTNU and The Norwegian Educational Loan Fund, through Quota Program. Their supports have been highly appreciated.

8. Summary References As in CO2 transcritical cycle air conditioning, there will be an optimum condition for a CO2 combined air conditioning and tap water heating system at which the system reaches the highest cooling-COP. The optimum condition is determined by components parameters such as gas coolers configuration and percentage of heat recovery. In the parallel configuration, there is an improvement in air conditioning side as heat recovery increased and the optimum condition will be depending on the percentage of heat recovery. However, in the series configuration the influence of heat recovery on the performance of the air conditioning side is insignificant. Total-COP as a representation of total efficiency of the combined system is higher compared to efficiency of an air conditioning system without heat recovery. As energy

[1] G. Lorentzen, Revival of carbon dioxide as refrigerant, International Journal of Refrigeration 17 (3) (1994) 292. [2] J. Pettersen, R. Aarlien, Progress in CO2 Vapour Compression System, in: Proceedings of the Oji International Seminar, Tomakomai City, Hokkaido, Japan, 1997. [3] P. Nekså, H. Rekstad, G.R. Zakeri, P.A. Schiefloe, CO2 -heat pump water heater: characteristic, system design and experimental results, International Journal of Refrigeration 21 (3) (1998) 172–179. [4] R. Aarlien, P.E. Frivik, Comparison of practical performance between CO2 And R-22 reversible heat pumps for residential use, in: Proceedings of the Natural Working Fluid’98, IIR-Gustav Lorentzen Conference, Oslo, Norway, 1998. [5] G.R. Zakeri, P. Nekså, H. Rekstad, Work Package 11: Prototype System Design and Construction Final Report 01/99, Energy Efficient And Environmental Friendly Heat Pumping Systems Using CO2 As Working Fluid, COHEPS, 1999.