1、NA-04-5-5 Field Performance of HPWH Preheating for Water Heating Systems in Schools David R. Dinse, P.E. Member ASHRAE Hugh I. Henderson, Jr., P.E. Member ASHRAE Member ASHRAE John O. Richardson, Jr., P.E. ABSTRACT This paper presents Jield-monitored performance data from a school where a heatpump w
2、ater heater (HPWH) was installed with a storage tank to preheat water for a gas-$red hot water system with multiple recirculation loops. The results demonstrated the efectiveness of HPWHpreheating for high water use areas in the school and conjirrned the high thermal losses from the recirculation lo
3、op system. The authors discuss the implications of this work as it relates to the overall design strategy for service water heating systems in schools. INTRODUCTION The use of heat pump water heaters (HPWHs) for commercial service water-heating (SWH) loads has been investigated and documented by a n
4、umber of field demonstra- tions. Those results have been summarized in an application guide (EPRI 1990) that recommends the best approaches for dcsigners to apply HPWHs in commercial buildings. The guide primarily addresses air-to-water HPWH systems, which use a refrigeration cycle to extract heat f
5、rom the air to heat water. Air-source HPWHs are especially attractive in commercial applications because space cooling is provided at no cost as by-product of water heating. Since commercial buildings are ofien cooling-dominated, the free cooling provided by the HPWH directly decreases energy use of
6、 the space cooling equipment. Because the efficiency of HPWHs is highest when heat- ing low-temperature water, these systems are often best applied as preheating systems that heat water to 110F to 120F (43C to 49C). While HPWHs are three to four times more efficient than electric resistance heating
7、elements, the heating capacity of these units is usually limited. Therefore, HPWHs are typically coupled with large storage tanks so that the HPWH runs for a significant portion of the day to heat the water volume. In some cases HPWH operation could be limited to off-peak utility periods when electr
8、icity costs are low. If the system is used to preheat water to more modest temperatures of 100F to 110F (38C to 43“C), then the HPWH system is even more efficient. This paper presents measured results from a school where a geothermal HPWH was added to preheat water for a gas- fired, centralized, wat
9、er-heating system. The results from this site confirm the benefits of an HPWH preheating system and confirm the substantial thermal losses from a large centralized recirculation loop. Standby losses from the water heater tank/ recirculation loop represented 37% of the annual water-heat- ing energy u
10、se. The paper then applies the lessons from this site and other past field studies to propose more optimized water-heating systems that are appropriate for modern school applications. FIELD TEST RESULTS FROM AN HPWH PREHEATING SYSTEM HPWH Field Test Site Description A geothermal HPWH system was inst
11、alled during construction at a large high school in Erwin, Tennessee (Hend- erson 2003). The 135,000 fi (12,540 m2) high school was constructed in 1999 with geothermal water-source heat pumps for space heating and cooling. The school has 38 classrooms with a student capacity of 900. A water-to-water
12、 heat pump was connected to the geothermal loop to preheat service hot water for the facility. The geothermal HPWH was originally installed as a preheating system for the gas-fired water heater David R. Dinse is a project manager and John O. Richardson, Jr., is a senior manager in energy use and ind
13、ustrial ecology technologies at the Tennessee Valley Authority in Chattanooga, TN. Hugh I. Henderson, Jr., is a principal at CDH Energy Corp. in Cazenovia, NY. 690 02004 ASHRAE. Cold Wahrr : Heat ;+GE pump -+-+ Water + HeateL * v Ground Loop - 750gal i Storage : Tank HPWH Preheat System TMWE Equipme
14、nt Heat Pump Water Heater Comments Nominal 5 ton 17.5 kW water-to-water heat pump -+ _ HPWH Storage Tank HPWH Circulation Pump 250 gal Water Heater e FGWH! -r 750 gallon 2840 11 capacity !A hp, cycles on and off with HPWH 1 Hot Water Loop Circulation Pumps Fixtures 1/12 hp, operate continuously year
15、-round (140F 60C loop pump never operated) Figure 1 Schematic of modrfed water heating system (with measured points shown). 250 gallon 950 LI, 1.4 MMBTUh 410 kW input 83% efficient (nameplate) (GWH) by diverting makeup water for the GWH system through the HPWH on a one-pass configuration. The origin
16、al design did not have a dedicated preheat tank associated with the HPWH. This system served the school cafeteria kitchen and a portion of the bathrooms. Locker room showers and other bathrooms are served by a second water-heating system not included in this field test. In 2001 the HPWH was reconfig
17、ured with a 750-gallon (2840 L) preheat tank and a circulation pump so that the HPWH capacity could continuously provide heat to the preheat tank. The performance improvements in the reconfig- ured system confirmed the necessity of including a dedicated preheat tank in HPWH systems (EPRI 1990). The
18、service water-heating system has both high-temperature (140F 6O“C) and low-temperature (120F 49“C) circulating loops that serve a portion of the school. The 140F (60C) water is used in the kitchen area, while the 120F (49C) loop serves nearby restrooms and hand wash sinks in the kitchen area. Figure
19、 1 schematically shows the modified water-heating system installed at the school. Table 1 lists the major compo- nents in the HPWH preheat system. The system preheats water entering the GWH to about 120F (49C). Data Collection and Instrumentation A datalogger was installed to continuously collect da
20、ta records at 15-minute intervals. The datalogger scanned the channels each second and averaged or totaled the data into 15- minute intervals. The data logger and the associated instru- mentation are described in detail in Henderson (2003). Table 2 summarizes the monitored data points, and Figure 1
21、sche- matically shows the location of each measured data point in the system. Field Test Results Data collection started in November 2002 and continued until March 2003. Over the four-and-a-half-month monitoring period, the water-heating system operated in three distinct modes: ASHRAE Transactions:
22、Symposia 691 O: 2: 4: 6: 8: 10: 12: 14: 16: 18: 20: 22: O: Data Point Description TMWE Entering Water Temperature 17 18 Units “F Figure 2 Operation of the water heating system on a typical school day. THPWE THPWL . Hot Water Entering HPWH Temperature Hot Water Leaving HPWH Temperature “F “F GWH Only
23、 (Mode 1). This established the base case with the conventional gas water heater supplying the entire water-heating load and the HPWH disabled. HP WH Preheat-Thermostat Controlled (Mode 2). This mode allowed the HPWH to run whenever the ther- mostat in the 750 gallon tank called for heating. HP WH P
24、reheat-Thermostate Clock Controlled (Mode 3). This only allowed the HPWH to operate when the tank thermostat called for heating only after the building had essentially shut down for the day. This demand-limiting mode ensured that HPWH operation did not add to the building peak demand from 10 a.m. to
25、 4 p.m. each day. TWHL TWS2 TWR2 TGE Figure 2 shows the operating pattern for a typical day when the HPWH was operating with the GWH (Mode 2). The peak hot water consumption is typically 2 to 4 gpm (0.13 to 0.25 L/s) during the school day from 8 a.m. to 2 p.m. The heat pump usage also corresponds to
26、 this period, with a slight delay as the preheat tank is initially emptied. At night when there is no water use, the time between gas pulses provided an indica- tion of how much gas use was required to overcome standby losses during the unoccupied hours. Water use for this system was typically 800 g
27、allons (3030 liters) per day during a school day and never exceeded 1000 gallons (3790 litters) per day during the four-and-a-half- month period. The peak hot water draw in a 15-minute period rarely exceeded 60 gallons (230 liters) and was never observed to exceed 90 gallons (340 liters). On weekend
28、s there was typi- cally little or no water use. As is common with many central water-heating systems, the existing natural gas water heater was significantly over- sized for this application. The gas input rating is 1.4 million Leaving Water Temperature “F Hot Water Supply Temperature - 140F 6O“C Lo
29、op Hot Water Return Temperature - 140F 60C Loop Ground Loop Entering HPWH Tempera- ture “F “F “F FTW FGWH THPTWL I HPWH/Tank Leaving Water Temperature 1 OF 1 Tempering Line Water Flow GPm GasFlow Cflh WHP 1 OF I TWS1 Hot Water Supply Temperature - 120F I r49oc1 Loop Heat Pump Power kWhih I OF I TWRI
30、 1 Hot Water Return Temperature - 120F r49oci LOOD I OF I Ground Loop Leaving HPWH Tempera- TGL Jture FMW I Makeuo Water Flow I Gom I Btu/h (410 kW) for this 250-gallon (950 1) tank. Given this, the resulting recovery rate is on the order of 5 to 10 minutes. The gas water heater runtime never exceed
31、ed 20% for any 692 ASHRAE Transactions: Symposia Table 3. Overall Load Met by HPWH in Each Operational Mode Operational Test Configuration GWH Only (Mode 1) HPWH Preheat - Thermostat Controlled (Mode 2) HPWH Preheat - Tstat/Time Clock Controlled (Mode 3) Portion of Hot Water Load Met By: HPWH GWH 0%
32、 100% 57-66% 3443% 5 1-64% 36-49% Mode Gas WH Only HPWH Preheat 1 5-minute interval during the four-and-a-half-month moni- toring period (even when the HPWH was disabled for the testing). Similarly, the HPWH was oversized in that the heat pump never ran for more than five to six hours a day to prehe
33、at the tank. A heat pump one-half or one-third that size (i.e., 2 ton 7 kW), would still have sufficient time (i.e., 12 to 18 hours per day) to heat the tank. This holds true even if it were restricted to operate only during non-peak periods. The thermal losses associated with the gas water heater t
34、ank and recirculation loop were estimated to be around 6,000 Bhdh (I .8 kW), using a gas recovery efficiency of 75%. This equates to just over 700 therms (20.5 MWh) (gas input) per year to overcome standby losses for the tank and recirculation loop, or 37% ofthe total water-heating energy for the ye
35、ar. The 120F (49C) recirculation loop system operated 24 hours a day, 365 days a year at this school. The recirculation pump on the 140F (60C) loop did not operate during the test period. If the 140F (60C) loop pump had operated, the loss rate would have been even greater. This rate of heat loss is
36、similar to the loss rate of 8,500 Btdh (2.5 kW) determined for the hot water tank and recirculation loop at another Tennessee school (Hiller et al. 2002). Table 3 summarizes the portion of the total water-heating load met by the HPWH preheating system in each operational mode. The preheating system
37、provided more than half of the heating required. Limiting HPWH operation to the off-peak period provided only a modest impact on the portion of the load met by the HPWH preheating system. Table 4 shows that the HPWH preheating system as tested has lower operating costs than the centralized gas water
38、 heater (GWH), especially when the HPWH only operates during off- peak periods when electric energy costs are much lower. Off- peak operation eliminates a monthly peak demand charge of Average Electric Average HPWH Electric Delivery Gas Delivery Cost Cost ($/kWh) COP or GWH Eff Cost ($NMBtu) ($/MMBt
39、u) 0.75 $12.15 $0.103 3.5 $8.62 Figure 3 Water heating daily operating costs with water use. HPWH Preheat-Time Clock Installed $9 per kW; in this case, the demand charge triples the effective energy costs for a month. The cost of operating each system actually varies substan- tially with the daily w
40、ater-heating load. When there is no water use, each system still costs $2 to $3 per day to operate. This operation is required to overcome standby losses of the tanks and distribution system. Figure 3 shows that, when water-heat- ing loads are low, additional energy is also required to over- come th
41、ermal losses from the HPWH preheat tank. The system with HPWH preheating costs more to operate until daily use exceeds about 200 gallons (760 liters) per day. This result illustrates the benefit of scheduling the HPWH preheat- ing system to only operate before days with significant water- heating us
42、e. $0.033 4.5 $2.15 ASHRAE Transactions: Symposia 693 Annual Gas Use, Annual Electric Use Annual Operating therms (MWh) (kWh) costs GWH Only (Mode 1) 1913 (56.0) - $1,743 HPWH Preheat (Mode 2) 1193 (34.9) 6996 $1,909 HPWH Preheat with Timeclock 1059 (31.0) 5194 $1,156 (Mode 3) By using the informati
43、on in Figure 3, and assuming a mix of days with full, partial, and no water use, the annual impact of each operating mode can be determined. Because operating costs are slightly higher with the HPWH on low and no water use days, the HPWH operating continuously to provide preheating (Mode 2) actually
44、 results in slightly higher net operating costs. If the HPWH only operates during non-peak times, then the energy costs to operate the HPWH are much lower and the annual savings approach $600, or one-third of base case operating costs, using local Tennessee utility rates. Annual Savings ($166) $587
45、DISCUSSION AND IMPLICATIONS Hot Water Loads in High Schools The hot water usage data from this field study as well as Hiller et al. (2002) show relatively modest hot water loads for high schools. While in each case the measured water use only represents a portion of the total load, it implies that t
46、he historic expectation of high school water use may be no longer valid. Traditionally the two largest DHW loads in schools were considered to be the kitchen and the locker rooms (for show- ers). Data in the ASHRAE Handbook-Applications (ASHRAE 2003) provides a design basis for water-heating loads f
47、or high schools. Table 6 in Chapter 49 states that aver- age water use rates are about 1.8 gallons (6.8 liters) per student per day for a high school and 0.6 gallon (2.3 liters) per student per day for an elementary school. The Handbook describes the difference between the two types of schools as be
48、ing primarily attributable to showers in the locker rooms. Applying these rules of thumb to the Erwin, Tenn., school implies that the overall average water use for its 900 students should be 1620 gallons (61 30 liters) per day, and about two-thirds of this water use or 1,000 gallons (3790 liters) sh
49、ould occur in the locker room showers. Similarly, peak hourly per student water use rates are 1 gallon per hour in the high school and 0.6 gallon (2.3 liters) per hour for an elementary school. For a student body of 900 these peak usage rates translate to 15 and 9 gpm (0.57 and 0.95 L/s), respectively, for high schools and elemen- tary schools. The most significant change in school water use may be in gym locker room usage. Many school maintenance workers now report that shower use for gym classes during the school day is very limi
