ENERGY RATING OF DOMESTIC WATER HEATERS ANZSES 1992, Annual Conference Darwin. G.L.Morrison and H.N.Tran School of Mechanical and Manufacturing Engineering University of New South Wales ABSTRACT A unified procedure is presented for rating of all forms of water heaters for energy efficiency and CO 2 pollution. The scheme provides a means of rating electric, gas, solar and heat pump water heaters. The procedure is based on data available under existing Australian standards and uses a computer simulation model to determine annual task cycle energy efficiency. It is argued that a ranking scheme based on metered energy consumed gives the best indication of both energy and CO 2 ranking of domestic water heaters. INTRODUCTION Energy efficiency rating of domestic appliances is usually evaluated for a standardisation set of operating conditions. This approach has been adopted for gas water heaters and is currently being considered for electric water heaters. However standardised rating procedures cannot be applied to solar water heaters or heat pump water heaters as these devices experience an extremely wide range of operating conditions and it is not possible to develop a meaningful ranking order from single point tests. One of the major impediments to the acceptance of solar and heat pump water heaters is the difficulty a consumer has in comparing the performance star rating scheme for a gas water heater with a claim of a level of annual energy saving for a solar water heater. If a different rating system is developed for electric water heaters (currently under consideration by Standards Australia) it will complicate the task of promoting high quality, low pollution solar and heat pump water heaters. This project has addressed the problem of rating conventional and renewable energy domestic water heaters and represents a unified scheme for rating all forms of water heaters. Energy efficiency rating of conventional heaters The existing star rating scheme for gas water heaters was developed as an internal gas industry means of differentiating between "standard" and "high efficiency" gas storage water heaters. The scheme provides a simple means of separating "high efficiency" heaters (usually with return flue or condensing flue), from conventional chimney flue systems. The gas-rating scheme does not take account of the effects of load cycle operation nor does it account for variations of environmental conditions such as cold water supply temperature and ambient temperature. A procedure for determining the energy consumption of electric water heaters has been presented in Australian Standard AS16.4 [1]. For off-peak water heaters (the most common electric water heater in eastern Australian states) AS16.4 uses a simple 1% adjustment to account for the varying temperature conditions in off-peak storage tanks, and makes no allowance for the effect of load induced mixing on the quantity of useable hot water, effect of flow diffusers to promote stratification or the effect of improved insulation on the promotion of stratification. Energy efficiency rating of solar water heaters The annual energy saving of a solar water heater can be evaluated by experimental procedures defined in Australian Standards AS2813 and AS2984 [2,3]. System performance evaluation procedures such as AS2813, using a solar simulator, or AS2984 using outdoor measurements, have the advantage that the systems are tested as a black box and hence the testing authority does not need to take the system apart, or need to know the
ENERGY RATING OF DOMESTIC WATER HEATERS 2 internal operation in order to determine the performance. The disadvantage of these procedures is that they are expensive to carry out or require a long monitoring period. The energy efficiency of a solar water heater depends on the characteristics of the system and on climatic conditions. Methods for quantifying the efficiency of solar water heaters include a) Solar simulation tests (Australian Standard AS2813) b) Short term outdoor tests (Australian Standard AS2984) c) Long term outdoor tests d) Outdoor comparative performance testing relative to a reference system e) Analysis of long term performance based on component testing of the solar collector and storage tank. Although the first two methods are well developed, both are expensive to perform. Long term outdoor testing requires monitoring over at least twelve months and outdoor comparative testing has significant uncertainty due to aging of the reference system. The solar simulator based standard (AS2813) was adopted to allow assessment of performance under controlled conditions. This standard also includes a procedure for computing annual energy savings for outdoor operation. Similar standards have been adopted in the USA [4] and by the International Standards Association []. However these standards only rate systems for a defined solar day. To provide the extra information needed to determine annual outdoor performance the Australian simulator standard requires a much longer test period and hence is more expensive to perform than the ASHRAE or ISO standards [4,]. An outdoor test procedure (AS2984) has also been adopted by Standards Australia. The test period is governed by climatic conditions at the test site and a test period of 9 to 12 weeks is generally required. The annual energy savings are determined using a model to correlate the short term tests. The existing Australian Standards can be applied to any design of solar water heater, however these methods are expensive to apply due to the high cost of simulator testing and the long test period required for outdoor testing. The performance of many types of solar water heaters have also been evaluated using correlation models (eg Fchart or Sunbear), however, correlation methods are not as accurate as detailed simulation calculations. Correlation modelling was developed to save computation time - with a corresponding loss of accuracy - however with the ready availability of fast desk top computers correlation models are being displaced by the more fundamental and hence more accurate detailed simulation models. STANDARDISED PROCEDURE FOR LOAD CYCLE EFFICIENCY RATING To account for the load cycle operating conditions imposed on all forms of water heaters, a procedure has been developed as part of this project that combines component testing and a mathematical simulation of load cycle operating conditions. As part of the simulation procedure, seasonal cold water and ambient temperature variations and user load patterns are defined. The annual load cycle energy consumption is determined by a short time step simulation of the temperature conditions in the storage tank and operation of energy input systems (electricity, gas or solar). This procedure is currently being considered by Standards Australia as a unified rating procedure for all forms of water heaters. Due to the high cost of load cycle testing of water heaters, calculation methods must be used to determine the annual energy use. The load cycle simulation procedure must account for daily and seasonal variations of load, temperature gradients in the water tank, flow induced mixing in the tank, the variation of losses as a result of limited recovery rate of the energy source and the efficiency of the energy source. For off peak electric water heaters the simulation procedure developed for this project correctly accounts for varying tank heat losses during the day and hence removes the need for the arbitrary adjustment factors incorporated in the current procedure in AS16.4. The procedure also accounts for the effect of load induced
ENERGY RATING OF DOMESTIC WATER HEATERS 3 mixing in off peak electric tanks and hence can be used to quantify load delivery capability as a function of user load patterns. Mathematical evaluation of the performance of a solar water heater is more complicated than for a conventional water heater because a short term time step analysis must be carried out in order to account for the transient characteristics of solar radiation. WATER HEATER LOAD CYCLE PERFORMANCE SIMULATIONS The annual performance of conventional water heaters operating under load cycle conditions can be readily evaluated from knowledge of the standing heat loss (or maintenance rate of a gas storage water heater), and draw-down delivery capacity of the tank. Once these factors are known (using existing standard test procedures) a short time step simulation program can be used to follow the temperature variations in the tank in response to loads, tank heat loss and energy source operation. The long term performance of solar water heaters can be readily evaluated if the efficiency of the solar collector and heat loss characteristics of the tank and solar plumbing are known. A detailed mathematical analysis over a typical climatic year for the location of interest will require inputs of component characteristics and hourly solar radiation and ambient temperature data. Appropriate climatic data is available for 22 locations in Australia including all capital cities [7]. Numerous simulation programs have been developed for determining the performance of solar water heaters. The most widely used computer model is the TRNSYS package [8]. Researchers in Australia have contributed to the development of this code and have also developed many extensions to suit Australian solar equipment and Australian component testing procedures [9]. TRNSYS has also been used extensively by Australian companies for developing new solar products and evaluating system designs. STANDARDISED RATING CONDITIONS Hourly load patterns The hourly draw off patterns in a domestic situation have been investigated in a number of studies [1,11]. The load patterns adopted by Standards Australia for solar water heaters and the pattern resulting from a study of water heaters in Victoria are given in Table 1. Table 1 Hourly Load Patterns for Annual Task Performance Evaluation Percent of daily load Time AS2984 VSEC 3 7 1 8 1 1 9 3 1 11 1 13 1 1 12. 12 16 12. 17 12. 18 12. 19 2 2 1 22
ENERGY RATING OF DOMESTIC WATER HEATERS 4 Seasonal loads The seasonal energy load patterns have been investigated for a number of locations in Australia. The two most detailed studies were by the University of NSW [12] and the Victorian Solar Energy Council [11]. The load patterns from these studies are shown in Table 2. The University of NSW load pattern was adopted by Standards Australia for AS2984. In the gas water heater energy rating scheme [13] the seasonal load pattern influence is neglected and a constant load is assumed to apply throughout the year. Peak winter load Table 2 Load Patterns for Annual Task Performance Evaluation Month Fraction of nominal load Univ of NSW VSEC January.7.68 February.8.68 March.8.74 April.9.81 May.9.91 June 1..92 July 1. 1. August 1. 1. September 1..92 October.9.91 November.9.81 December.8.74 The peak winter daily energy load in a domestic situation varies with the location. The only published studies are for Sydney [12] and Melbourne and northern Victoria [11]. The peak load reported in these studies ranged form 37 MJ/day in Sydney to 42 MJ/day in northern Victoria. A range of typical peak loads were determined for different zones (see section.) in Australia as shown in Table 3. Table 3 Peak winter daily energy loads Zone Large system peak load MJ/day Small system peak load MJ/day 1 3 18 2 3 18 3 37.7* 22. 4 37.7* 22. Weather conditions The annual load cycle performance of conventional water heaters is a function of ambient temperature and user demand. The performance of solar and solar boosted heat pump water heaters is also a function of solar radiation, wind and cloud cover conditions for the location of interest. Weather data required for inputs to a simulation model are available in the Australian Typical Meteorological Year data bank. To minimise the amount of climatic data required four weather zones were adopted by Standards Australia in AS2984 (Fig. 1). Cold water temperature data for each climate zone in Australia has been defined in AS2984. Fig 1 Climate zones (AS2984)
ENERGY RATING OF DOMESTIC WATER HEATERS PERFORMANCE EVALUATION PROCEDURE As the procedure in the rating scheme requires a mathematical model for assessing annual task performance the application is restricted by the availability of suitable models. Currently computer simulation models are available for the following system components. Conventional gas storage water heaters Conventional electric storage water heaters with any tariff Single tank solar water heaters with electric or gas in-tank boosting. Multiple tank solar water heaters with separate solar and auxiliary tanks. Heat pump water heaters with the evaporator exposed to solar irradiation. The following solar water heater configurations are included Solar water heaters with flat plate, concentrating or evacuated solar collectors. Thermosyphon or pumped fluid circulation through the solar collectors. Annular tank in tank heat exchanger in the thermosyphon loop. Horizontal or vertical water storage tanks. Water heater task performance simulation model The task performance calculations are based on the TRNSYS [8] simulation program. TRNSYS is a transient system simulation program developed at the University of Wisconsin Solar Energy Laboratory. TRNSYS has gained worldwide acceptance as a result of its modular construction and due to the availability of the code in the public domain. The modular nature of the code means that it can be readily extended to include new energy devices and systems. To develop a simulation package suitable for Australian solar water heating products the Solar Thermal Energy Laboratory at the University of New South Wales has extended the TRNSYS code to include many of the unique features that have been developed by Australian water heater manufacturers. The modified code is referred to as TRNAUS. RESULTS Typical annual task cycle energy consumption and CO 2 generation for a solar water heater is shown in table 4 Table 4 Typical load cycle performance of a solar water heater (Climate zone 3) Radiation Ambient Load Volume Tank Aux Tout CO 2 temp loss produced MJ/m 2 C MJ/d L/d MJ/d MJ/d C kg/day JAN 21.4 23.1 28. 138 11.8 4.3 68.8 1.6 FEB 2.6 22.3 32. 169 1.7 6.3 66.3 2.3 MAR 18.8 22.4 34. 186 1.1 9.4 64.8 3.4 APR 16.7 19. 36. 28 9.7 14.4 61.8.2 MAY 13.4 14.6 38. 223 9.8 24.7 9.6 9. JUN 12.9 12.6 4. 221 1.2 27.6 9.7 1. JUL 1.7 11.3 4. 214 11.1 23.1 9.9 8.4 AUG 17.2 13.6 4. 27 1.7 18.9 6.8 6.9 SEP 19.4 16.7 4. 23 1.7 14.2 62.3.2 OCT 21.9 16.8 38. 189 11.7 9.8 64.6 3.6 NOV 2. 19.8 36. 186 1.6 8.4 64.4 3. DEC 23. 22.3 32. 14 12.1 2.9 68.7 1.1 Annual Load = 1326 MJ, Auxiliary = 3 MJ, CO 2 = 1816 kg
ENERGY RATING OF DOMESTIC WATER HEATERS 6 Radiation = Irradiation on the collector slope Ambient temp = Ambient temperature (24 hour average) Load = Useful energy delivered Volume = Volume of water delivered Tank loss = Heat loss from solar components of the system Aux = Auxiliary energy used by boost system (includes booster efficiency) Tout = Average hot water outlet temperature Comparison of water heater performance The metered energy consumption of seven types of domestic water heaters is shown in Figure 2. All systems were assessed for the same load conditions (4 MJ/day peak winter load) for zone 3 weather conditions. The annual energy delivery efficiency ranged from approximately 2% for the three solar systems to 4% for the standard gas storage water heater. The annual energy efficiency of the "high efficiency" gas system was % and approximately 8% for the two conventional electric systems. The comparative performance information in Fig 2 clearly differentiates between different products and if presented as a 1 point star rating system (with two stars reserved for future developments), the following ranking would apply (stars = 1-(annual energy GJ)/3 ) Ranking System (7 to 8) Solar water heaters (electric boosting) and solar boosted heat pumps (6 to7) Solar water heaters (in-tank gas boosting) (4 to ) Electric storage water heater (2 to 3) High efficiency gas (1 to 2) Standard gas However, the gas industry objects to such a ranking scheme as gas is supplied as primary energy, whereas the electric systems use energy supplied via a low efficiency power conversion system. The most useful way to rank systems with respect to a measure of primary energy is to use the level of C 2 produced (or equivalent for gas leakage). The results of such an analysis are shown in Figure 3. The differentiation between the systems is more noticeable than the difference in metered energy consumption (Fig.2). On a 1 point scale the following ranking would apply (stars =1 -(Annual CO 2 tonnes)/.6 ) Ranking System (7 to 8) Solar with gas boosting (6 to 7) Solar with electric boosting and solar boosted heat pumps ( to 6) High efficiency gas (4 to ) Standard gas (1 to 2) Electric storage water heaters The relative ranking of systems may vary slightly between products in the one category and will vary with climate for the solar and heat pump systems. A similar set of energy and CO 2 results for Darwin is shown in figs 4 and. For the high radiation and lower load conditions applicable to Darwin, the heat pump system is not as effective as the direct solar systems. However, for higher loads the heat pump system performance would improve while the direct solar system performance will deteriorate. The variation of performance with location is shown in figs 6 and 7. The interesting feature of these results is that the relative ranking of each system does not change significantly with location (within the limits of a 1 point ranking) hence a single rating scheme could be developed for Australian-wide application. Choice of ranking scheme The reason that separate rating schemes have been developed for gas and electric water heaters is the lack of agreement between gas and electric supply authorities with regard to the selection of primary or metered energy as the ranking variable. The use of primary energy ranking is often presented as a valid measure of the pollution ranking of conventional gas and electric water heaters, however, this may not be correct if the life cycle efficiency of both products is considered. The efficiency of a 2 year old gas water heater is significantly
ENERGY RATING OF DOMESTIC WATER HEATERS 7 less than for a new gas heater while electric water heaters show only small variation of heat loss with age. If the average life cycle efficiency was used to rate gas water heaters the relative ranking of gas and electric water heaters would not significantly influenced by the choice of primary or metered energy as the analysis parameter. In addition to the lower thermal efficiency of old gas water heaters (hence increased CO 2 production) the generation of SO 2 and NO X also increases due to incomplete combustion in dirty burners. A further disadvantage of gas water heaters is that the SO 2 and NO X production is distributed in the urban areas and can only be controlled by expensive maintenance and regular replacement of the burners, whereas the pollution produced in electric power plants is concentrated at a few locations and if society decides to reduce atmospheric pollution it is relatively easy (though expensive) to install pollution control systems at power stations. If a decision were made to reduce urban atmospheric pollution it would be cheaper to promote electric water heaters and to install pollution controls at the power stations rather than replace all gas water heaters with low pollution burner systems. As a result of these considerations a ranking scheme based on metered energy use is to be preferred. REFERENCES 1. AS16.4 Storage water heaters, Part 4: Calculations of energy consumption 2. AS2813 Solar Water Heaters- Method of Test for Thermal Performance - Simulator Method 198 3. AS2984 Solar Water Heaters -Method of Test for Thermal Performance - Outdoor Test Method, l987. 4. ASHRAE 9 Methods of Testing to Determine the Thermal Performance of Solar Domestic Water Heating Systems. l981. International Standards Association. DP3949-1, Solar Heating: Domestic Water Heating Systems Part 1: Performance Rating using Indoor Test Methods 6. AS23, Glazed flat plate solar collectors with water as the heat transfer fluid. Method of testing for thermal performance, 1982 7. Condensed Solar Radiation Data Base for Australia. Morrison, G.L. & Litvak, A. School of Mechanical & Industrial Engineering, University of NSW Report l988/fmt/1. 8. TRNSYS 13 User Manual, Kline, S.A. et al. University of Wisconsin Solar Energy Laboratory,l99. 9. TRNSYS Extensions for Australian Solar Products (TRNAUS). Morrison, G.L. School of Mechanical & Manufacturing Engineering, University of NSW. Report l991/fmt/2. 1. Lacey,J.C Measurement of solar water heater performance. Electricity Supply Association conference 1976 11. Guthrie,K.I and Kimpton,N.C. Pilot study to define the patterns and quantities domestic hot water consumption in Victoria. NERDDP project No 817, final report 1987 12. Morrison,G.L and Sapsford,C.M. Performance of Thermosyphon Solar Water Heaters - Final Results, Report 1982/FMT/1 (Kensington, University of New South Wales) 13. Australian Gas Association Standard AG12
ENERGY RATING OF DOMESTIC WATER HEATERS 8 3 2 Purchased energy GJ/year 2 1 1 Electric offpeak Electric 24hr Gas (standard) Gas (high efficiency) Solar + electricity Heat pump (solar) Fig 2. Annual task cycle energy consumption, Zone 3, peak load 4 MJ/day. 6 CO2 production Tonnes/year 4 3 2 1 Electric offpeak Electric 24hr Gas (standard) Gas (high efficiency) Solar + electricity Heat pump (solar) Fig 3. Annual system CO 2 production, Zone 3, peak load 4 MJ/day.
ENERGY RATING OF DOMESTIC WATER HEATERS 9 2 2 Purchased energy GJ/year 1 1 Electric 24hr Gas (standard) Gas (high efficiency) Solar + electricity Heat pump (solar) Fig 4 Annual task cycle energy consumption, Zone 1, peak load = 3 MJ/day. CO2 production Tonnes/year 4 3 2 1 Electric 24hr Gas (standard) Gas (high efficiency) Solar + electricity Heat pump (solar) Fig Annual system CO2 production, zone 1, peal load = 3 MJ/day.
ENERGY RATING OF DOMESTIC WATER HEATERS 1 3 Gas (standard) 2 Purchased energy GJ/year 2 1 1 Solar + electric Heat pump + solar Electric 24hr 1 3 4 1 3 4 1 3 4 1 3 4 1 3 4 Climate zone Fig 6. Annual energy versus climate zone, peak laod = 3 MJ/day for zone 1 and 4 MJ/day for zones 3 & 4. 6 Electric 24hr CO2 production Tonnes/year 4 3 2 1 Solar + electric Heat pump + solar Gas (standard) 1 3 4 1 3 4 1 3 4 1 3 4 1 3 4 Climate zone Fig 7. Annual CO2 produced versus climate zone, peak load = 3 MJ/day for zone 1 and 4 MJ/day for zones 3 & 4.