1、NA-04-5-2 Measured Performance and Impacts of “Drop-In” Residential Heat Pump Water Heaters John J. Tomlinson Member ASHRAE ABSTRACT This paper presents final results of a jeld evaluation of residential heat pump water heaters. This heavily instru- mented jeld study was conducted in I7 homes across
2、the United States over a period of 18 months. This study was unique in that the heatpump water heaters could be operated as electric resistance water heaters or as designed, heatpump water heaters with resistance backup. Field measurements of hot water usage, energy consumption, temperatures, and co
3、ndensate generation provided ample data to assess the performance and ejiciency of the heatpump water heater in a range of settings. The paperpresents comparative results on energy eficiencx dehumidi3cation performance, electric demand, and loadfactor impacts important to utilities as well as measur
4、ed customer impacts in a switchover from electric resistance water heaters to heat pump water heaters. INTRODUCTION Water heating accounts for 12% of all of the energy used in buildings, and buildings account for one-third of all energy used in the nation (BTS 2000). Consequently, improving the effi
5、ciency of water heating can play a significant role in reduc- ing the nations thirst for energy. The market for residential water heaters is about evenly split between electric resistance and gas across the nation; however, there are many states where electric resistance water heater sales far outnu
6、mber gas water heaters sales. In Florida, for example, 85% of all water heaters sold are the electric resistance type. The efficiency of electric resistance water heaters has just about topped out, and the efficiency market is tightly compressed: newly enacted efficiency standards for electric water
7、 heaters sold beginning January 2004 are only 4% higher Richard W. Murphy, Ph.D. than previous efficiency standards for electric water heaters. On that date, the least efficient 50-gallon electric resistance water heater had an energy factor of 0.90, as compared to 0.86 before then, and the most eff
8、icient are probably in the range of 0.94. There is simply not much room lefl for further improve- ment in the efficiency of electric resistance water heaters. The heat pump water heater (HPWH), however, can provide a quantum leap in efficiency. Like an air conditioner or refrigerator, the HPWH emplo
9、ys a vapor compression refrigeration cycle to transfer heat from the air surrounding the water tank into the water. This can produce useful cooling and dehumidification of the air as well as providing hot water at high efficiency. There are two basic designs of residential HPWHs. The add-on type, co
10、mposed of a compressor, air-to-refrigerant evaporator, controls, and a water-cooled condenser, is installed in conjunction with an existing storage water heater. The add-on type also contains a small pump to circulate water from the existing storage water heater to the HPWH when the tank needs to be
11、 heated. In an add-on HPWH installation, piping is installed between the existing storage water heater and the HPWH, and the HPWH is wired so that the HPWH essentially replaces the function of the tanks lower element. Being retrofittable, the add-on HPWH allows the customer to retain the storage wat
12、er heater. On the other hand, the integral HPWH is a single package consisting of the HPWH compo- nents as well as the storage tank. Through the combined research efforts of several organizations, a “drop-in” version of the integral HPWH has been developed, as shown in Figure I. Termed “drop-in,” th
13、e design is intended to target the large replacement market for residential electric resistance storage water heaters. The drop-in design is unique in a number of J.J. Tomlinson is leader of the Building Equipment Group and R.W. Murphy is a research and development engineer in the Engineering Scienc
14、e and Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tenn. 664 02004 ASHRAE. Cornmessor. Removable Evaporator, Fan, and Controls Inside Wrapped Coil Foam Figure 1 Drop-in residential heat pump water ?teutez ways: it uses a small (-4000 Btu/h) hermetic reciprocating compressor, a cond
15、enser consisting of tubing tightly wrapped around the lower portion of the steel tank, and upper and lower heating elements that are controlled as backup during times of heavy draws or unfavorable compressor operating conditions. THE FIELD STUDYIEVALUATION Following laboratory development of the dro
16、p-in HPWH design, we instrumented 17 of the drop-in HPWHs and initi- ated an 18-month field study in conjunction with ten utilities around the nation (Murphy and Tomlinson 2002). instrumentation and Data Collection Instrumentation points and sensors were installed into each HPWH, as shown in Figure
17、2, prior to shipment to each test site. We used the following sensors and instrumentation locations: Type-T thermocouples (TC) to measure compressor suc- tion and discharge temperatures, inlet and outlet water temperatures, evaporator and ambient temperature, as well as tank temperatures at six unif
18、ormly spaced axial locations inside the tank. Watt meters to measure compressor, upper and lower element, and evaporator fan power. Flowmeter to measure hot water flow (draws) through the tank. Special flow instrumentation to measure rate of conden- sate production at the evaporator of the HPWH. Moi
19、sture sensor to measure ambient relative humidity. Potentiometer to measure the thermostat setting of the HPWH. TC?1 TC10 Flowmeter Hot water 7 out Evapmior . TC3 Condensate volume owrate TCl2 = Ambient Temp RH =Relative HumW DCV =Thermostat Setting WATTS Figure 2 Instrumentation points in HPWH fiel
20、d evaluation. We devised and used a data collection strategy that recorded data at intervals that were appropriate for the type of data to be collected: Ten-minute scans of ambient, compressor suction, and compressor discharge temperatures; relative humidity and thermostat setting Thirty-second, eve
21、nt-triggered scans of component elec- trical power, tank temperatures, and condensate produc- tion Two-second, event-triggered scans of inlet and outlet water temperatures and water flow rate Data were collected and downloaded from each site on a weekly basis through a modem and phone line. In addit
22、ion to data collection, we also had control over each unit so that we could operate it in two modes: as an HPWH with resistance backup (as-designed mode) or as a conventional electric resis- tance water heater. Our evaluation plan called for operating each unit for periods throughout the field study
23、 in each mode to determine relative performance and energy savings. Site Selection House designs and sites (homes) selected for the study consisted of designs typical of each region (e.g., basement, crawlspace, slab-on-grade) as well as water heater location (e.g., garage, utility room). Table 1 lis
24、ts the locations and char- acteristics for the sites chosen. ASHRAE Transactions: Symposia 665 Table 1. Characteristics of HPWH Field Test Sites Unit State 1 Georgia 2 Georgia Surrounding Space: Residents: C = Conditioned A = Adult Source of S = Semi-Conditioned Previous Electric C = Children Water
25、Unit Location U = Unconditioned Water Heater 2 A, 2C Well Utility room C 50-gal resistance 2A. 2C Citv Basement U 50-gal resistance - 3 Florida 2A, 1C 4 S. Dakota 2A, 2C 5 Connecticut 2A 6 Florida 2A, 3C City Garage U None City Basement C 50- gal resistance City Basement S Add-on HPWH City Garage U
26、SO-gal resistance HPWH Performance The field study began at each site as the existing water heaters were replaced by drop-in HPWHs. Because of this staggered installation of the HPWHs, initiation of data collec- tion took place over a six-month period. Data from each site were collected using on-sit
27、e dataloggers, downloaded on a weekly basis, and data summaries for the week were prepared and analyzed. For most of the 1 %month period of the experi- ment, the units were operated in the “as-designe average weekly HW consumption. delivered efficiency of each unit operated in either mode. With unit
28、s operating in the resistance mode, we measured units average efficiency to be 0.86, and in the HPWH mode (as designed), we measured average efficiency to be 2.00. With the HPWH operating as designed, we determined energy savings by comparing actual electrical energy consumption with the electrical
29、energy that the unit would have used at the measured efficiency while in resistance mode operation. The results of this analysis for each of the 17 units are shown in Figure 5. These results show that for all units, the energy savings range from 4 I% for unit 16 located in a Washington basement to 6
30、2% for Georgia unit 1 located in the conditioned space. The weighted average energy savings was 55% from this field evaluation. As shown in Figure 5, unit 17 (Oregon basement location) showed a 54% energy savings despite its low COP (COP = 1.18). The reasons for the low COP with unit 17: cool year-r
31、ound ambient temperatures (48“F), high WH thermostat setpoint (1 45“F), and very low water use (I 64 gal/week), each of which contributes to high relative standby losses from the tank. Even when operated in the resistance mode, the COP of unit 17 was 0.54. Therefore, rather than an outlier, the data
32、 shown on unit 17 is real and accurately represents the perfor- mance of the HPWH under these conditions. Customer Impacts In any changeover to a new technology for heating, cool- ing, or water heating, there is the potential for customers to be impacted by the way that the new product performs. Cus
33、tom- ers are impacted by the delivery of hot water; therefore, it was important to evaluate this issue. The two-second, event-initi- ated data on hot water flow and temperature allowed us to define draws and to calculate information contained in them. We defined the initiation of a draw event as a d
34、ata record . . m 0.9 - 00 o 5 50 - . cbsb O0 32 W O however, the average draw appeared to be smaller. Defi- cits appeared to grow marginally during the wintertime due to cold ambients and cooler initial pipe temperatures. However, as with unit 8, there did not appear to be any relation between opera
35、ting mode (HPWH or resistance) and hot water deficits. Unit 1 was located fully in the conditioned space in an Atlanta house. The draw performance for this unit is shown in Figure 8. It can be seen that the average hot water draw was about one gallon and that the deficits were small. The fact that u
36、nit 1 is located inside the house, the house is in a warm climate, the HPWH is relatively close to all of the hot water fixtures in the 200 $ 160 P $, 60 40 x 8 120 no 25 20 3 U 15 8 10 zi i 5 O MAMJJASONDJFMAMJJ Month -Resistance Mode Interval * Drawslday o Gallonslday 0 Deficitslday Figure 7 Draw
37、and hot water (HW) dejcits-unit 16. 120 30 25 - . ._ . - . - . . - - ,100- - - m P O MAMJJASONDJFMAMJJ -Resistance mode interval Drawsfday o Gallonslday Defitsiday Month Figure 8 Draw and hot water (HW) deficits-unit 1. house, and the typical draws are small, suggests that deficits would be small, a
38、s they appeared to be from Figure 8. As before, there was no evidence that the HPWH retrofit has any effect on hot water deficits. Cooling and Dehumidification. Cooling and dehumidi- fication provided by the HPWHs were also determined. We calculated the net cooling from the HPWH from first princi- P
39、h Qc = AU -I- CQd Qloss -Ein , (1) where Qc = net cooling (cooling at the evaporator - compressor XQd Qloss = heat loss from the tank to ambient, Ein = electrical energy to the tank, and AU = intemal energy change of the tank. With the exception of Qloss, all four terms on the right side of Equation
40、 1 could be determined from field data. The change in tank internal energy (AV) was determined using the six tank temperature sensors, and the total energy in the hot shell heat loss), = total thermal energy in all draws from the tank, 668 ASHRAE Transactions: Symposia water removed from the tank (C
41、Qd) was determined from the draw analyses. Tank heat loss (Qloss) was determined using the tank heat loss coefficient measured in the laboratory prior to the field test and field-measured data on average tank temperature and ambient temperature. From these data, the net total cooling, Qc, was determ
42、ined. We measured dehumidifi- cation in the field using a custom device that determined the volume of condensate generated before it was sent down the drain. The results of these measurements and calculations for unit 8 in an unconditioned space in an Atlanta house are shown in Figure 9. The total c
43、ooling rises as summer turns into winter, principally due to cooler incoming water temperatures and cooler ambient temperatures. During the resistance mode intervals, there was no compressor operation and, conse- quently, no cooling produced. Condensate was produced at the rate shown for the summer,
44、 but little condensate was produced during the winter. Consequently, it appears that the HPWH in an unconditioned space helps to dehumidify the space during the summer. The differences in dehumidification produced in the same month for two successive years could be weather- related or caused by othe
45、r unmeasured factors. The results of cooling and dehumidification measure- ments for unit 1 in a conditioned space in an Atlanta house are shown in Figure 10. Total cooling produced by the HPWH increases from summer to winter, as in the case for unit 8. Interestingly, for most of the test period, HP
46、WH evaporator temperatures were not low enough to do much dehumidifica- tion, i.e., the house air-conditioning system accomplished the dehumidification. However, during the March-May period when the houses air conditioner was not being used, the HPWH did perform dehumidification as shown. It should
47、be pointed out, however, that although the HPWH accomplished essentially no dehumidification during the summer, it still provided cooling and reduced the total cooling load of the homes air-conditioning system by approximately 0.2 tons. We conclude from this that the HPWH in a conditioned space may
48、not accomplish dehumidification effectively; however, an HPWH located in an unconditioned or semi-conditioned space (e.g., a basement) would accomplish dehumidification. Figure 9 Cooling and dehurnidiation perforrnance- unit 8. UTILITY IMPACTS Electricity producers provide generation capaciy suffi-
49、cient to meet the aggregate electric demand from residential, commercial, and industrial customers. A key measure of the benefit of a technology, such as the HPWH, to utilities is the diversified demand-a measure of the average demand (kW) that the technology distributed to a large number of customers would impose on the utility. In the HPWH field study, the aver- age diversified electric demand for a subset of units operated in the HPWH mode for a period and in the resistance mode for another period was determined. This analysis was done to evaluate the impact of a
