Residential ground source heat pumps with integrated domestic hot water generation: performance results from long-term monitoring.
Ground source heat pumps (GSHPs) show great promise as a means to substantially reduce house energy consumption through the use of the soil between 6 and 600 feet below the surface of the earth as the thermal energy source or sink. In the years preceding this study, there was an observed rate of 12% annual increase in the number of installations primarily in the Northern Midwest and along the East Coast (Lund 2004). The majority of these installations were closed loop systems with vertical ground wells.
As energy used for heating and cooling decreases, the amount of energy used for potable hot water can become the largest single point of household energy consumption. One strategy to reduce energy consumption for hot water generation in an all-electric home is to use a desuperheater. Research performed by Merrigan and Parker (1990) indicates measured efficiency levels of 110% for desuperheaters installed on 1980s vintage air source heat pumps in Florida.
Through IBACOS' work under the U.S. Department of Energy's Building America Program, Pine Mountain Builders collaborated with IBACOS in 2008 to build two houses located in Pine Mountain, Georgia, that met the 50% source energy reduction level with respect to the 2008 Building America Benchmark definition (Hendron 2008). To accomplish this, each house uses a GSHP for heating and cooling with desuperheater assisted potable hot water generation. The houses were unoccupied through the course of the study, eliminating variability due to user behavior. The two goals of the project were to document the space conditioning and water heating efficiency of the GSHP systems installed in the houses.
These two homes both have four bedrooms, 4.5 baths, and the same usable interior area. However, House 1 is a single story, while House 2 is a two story and had slightly more usable interior floor space due to a loft on the second floor. Full specifications for the houses are shown in Table 1 (IBACOS 2010).
Table 1. Building Specifications Building Envelope Geometry 2,024 sq. ft., 1-story, or 2,946 sq. ft., 2-story, 4 bedrooms, 4.5 bathrooms Roof Unvented attic, spray foam on underside of roof sheathing R-30 Walls 2x6 wood framing @ 16" o/c open cell spray foam R-22 cavity Foundation 12" to 18" tall monolithic slab with 1" xps (R-5) slab edge insulation Windows U = 0.38, SHGC = 0.35, wood frame Infiltration ACH (natural) = < 0.25 Mechanical Systems Heating 4.2 COP (rated)* GSHP Cooling 20.1 EER (rated)*GSHP. House 1: 2.6 tons. House 2: 3.8 tons. Domestic Hot Water 40- r 85-gallon electric tank, 0.94 or 0.92 nominal EF, desuperheater Ducts In conditioned space Leakage 2% to outside Ventilation Passive into return with ventilation controller Appliances ENERGY STAR, all electric appliances Lighting 100% energy efficient fixtures -- Fluorescent and LED * Rated performance at ANSI/AHRI/ASHRAE/ISO 13256-1 (ASHRAE 2005) standard rating conditions for ground source heat pump applications
As part of the upgrades necessary to achieve 50% energy savings with respect to the 2008 Building America Benchmark, both houses use dual capacity GSHPs for their space heating and cooling. The GSHP used in House 1 is rated at 9.1 kW (2.6 tons), while the GSHP in House 2 is rated at 13.4 kW (3.8 tons). The ground heat exchanger consists of a closed loop in two (House 1) or three (House 2) vertical bores, each approximately 180 feet deep, using two 1-inch-diameter polyethylene pipes with a U-bend at the bottom end of each bore hole (IBACOS 2008).
In addition, the houses incorporate a desuperheater connected to an electric tank type water heater rated at 0.94 nominal energy factor (EF) and 151 L (40 gallon) capacity in House 1 and 0.92 nominal EF and 322 L (85 gallon) capacity in House 2. The desuperheater, or supplemental heat exchanger, transfers excess heat from the GSHP compressor to water in the domestic water heater (DWH) tank, improving its efficiency. Preliminary analysis indicated that the desuperheater could offset up to 80% of the domestic hot water electrical demand in the Summer.
The first goal of this study was to compare the field measured coefficient of performance (COP) with the manufacturer's listed values of the two GSHPs. The second goal was to compare the measured efficiency of the desuperheater assisted DWH system to other market available DWH systems.
The COP of the GSHP was determined using thermal energy metering on the ground loop and electricity submetering of the compressor, ground loop pump, and air handler unit (AHU) fan. All results were calculated using Equations 1, 2, and 3 below:
[COP.sub.GSHP] = [TE.sub.GroundLoop] - [TE.sub.Compressor]/[EE.sub.Compressor] + [EE.sub.GroundLoopPump] + [EE.sub.AorHandlerFan] (1)
[TE.sub.Compressor] = 0.3 * [EE.sub.Compressor] (2)
[TE.sub.GroundLoop] = ([T.sub.GroundLoopOut] - [T.sub.GroundLoopIn] x [Q.sub.GroundLoop] x [[rho].sub.Water] x [C.sub.P] (3)
[TE.sub.GroundLoop] is the thermal energy sent to (positive value) or extracted from (negative value) the ground loop wells
[TE.sub.Compressor] is the thermal energy given off by the compressor
[EE.sub.Compressor] is the site electrical consumption of the GSHP compressor
0.3 is the assumed fraction of electrical energy input to the compressor that is lost to the surroundings as heat
[EE.sub.GroundLoopPump] is the site electrical consumption of the ground loop pump
[EE.sub.AorHandlerFan] is the site electrical consumption of the air handler unit fan
[T.sub.GroundLoopOut] is the temperature of water leaving the GSHP and traveling toward the ground loop well
[T.sub.GroundLoopIn] is the temperature of water entering the GSHP (entering water temperature [EWT]) returning from the ground loop well
[Q.sub.GroundLoop] is the volume of water flowing through the ground loop
[[rho].sub.Water] is the density of water
[C.sub.P] is the specific heat of water
Thermal energy metering on the ground loop consisted of a high temperature, chemical resistant flow meter mounted on the loop line returning to the GSHP unit from the ground loop, with temperature measurements on both the outgoing and incoming fluid using Type T immersion thermocouple probes (Figure 1). Thermal energy metering was performed in the same manner to determine the energy output from the DWH to the fixtures in the house and on the circulation loop connecting the desuperheater in the GSHP to the DWH.
[FIGURE 1 OMITTED]
Electricity consumption of the ground loop pump, compressor, AHU fan, and desuperheater loop pump were each individually measured using pulse output Watt-hour meters (4Hz full scale output frequency) that were installed with appropriately sized current transducers (CTs) rated at no more than a factor of 2 greater than the anticipated maximum load. For circuits with only resistive loads (the electric resistance heating elements in the DWH), voltage output current transducers were installed and Watt-hours then calculated by multiplying the measured current by the typical line voltage determined by a onetime measurement.
The electrical measurements for the DWH system were taken in the main circuit panel, while measurements related to the performance of the GSHP were taken at the unit. Type T thermocouples were used for the indoor air temperature measurement in each thermostatically controlled zone; reference temperature was taken at the data logger using a thermistor. Outdoor temperature and relative humidity were also measured.
All sensors were connected to a central data logger using low voltage shielded wire, and 60Hz noise cancelling integration was used to mitigate any noise due to electrical power lines. Sensors were sampled every 10 seconds, and data were averaged on a minute and hourly basis. A cellular data modem was used to allow daily remote collection.
During the monitoring period of January 2010 to April 2011, the house was unoccupied, and interior thermostat setpoints were maintained at constant values of 20.6 [degrees] C (69 [degrees] F) heating and 23.3 [degrees] C (74 [degrees] F) cooling. In order to determine the efficiency of the desuperheater, a solenoid valve controlled by the central data logger was installed on the hot water spigot in the washing machine box in each house, and operated on a regular basis according to a simplified version of the schedule used in the Building America House Simulation Protocols (Hendron and Engebrecht 2010), as only one flow rate, 0.095 L/s (1.5 gpm), was available from the fixed orifice opening. Flow rate through the solenoid valve was fixed at and solenoid valve opening duration varied to simulate different volumes. In order to more accurately simulate the usage patterns of real people, three daily water consumption volumes were used during the weekly schedule: days 1 through 3 use 177 L (46.8 gallons) each day; for days 4 and 5, an additional 99.5 L (26.3 gallons) were used for a total of 277 L (73.1 gallons) per day; and on days 6 and 7, an additional 78.0 L (20.6 gallons) were consumed for a total of 355 L (93.7 gallons) per day. Typical flow events simulated the water used by showering, hand washing at a sink, and the operation of the dishwasher and the clothes washer. Individual flow events ranged from 1.9 L to 57 L (0.5 gallon to 15 gallons). Scheduled hot water consumption occurred for the duration of the monitoring period.
The efficiency (the unitless ratio of site fuel energy input to thermal energy output, EF) of a DWH is typically determined based on measured thermal energy output of the water heater and measured electrical input (Equation 4). However, in this study, the calculation of the efficiency (Equation 5) is somewhat more complicated because the thermal energy provided by the GSHP is not free; it is created from electricity proportionately to the COP of the GSHP. It should be noted that the desuperheater may affect the GSHP COP, but this impact was not measured.
EF = [TE.sub.output]/[E.sub.ER] (4)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
[TE.sub.desuperheater] is the measured thermal energy provided to the DWH from the GSHP
[TE.sub.output] is the measured thermal energy provided from the DWH to the house
[E.sub.ER] is the measured electricity supplied to the heating elements in the DWH
[COP.sub.GSHP] is the seasonal average measured COP of the GSHP
Space Conditioning Coefficient of Performance
Coefficient of performance values were calculated every 10 seconds and were averaged (weighted with respect to GSHP runtime) on a minute and hourly basis. It was assumed that during a given hour, the system operated in only one mode, heating or cooling. Based on hourly data outputs, monthly average COPs were created, weighted by runtime in each mode. The results plotted in Figure 2 are the COP for the mode in which the majority of the runtime occurred. Measured system cooling mode COPs were closer to the range of manufacturers' values (Table 2) than measured system heating mode COPs. Table 3 shows seasonal average system COPs for heating or cooling.
[FIGURE 2 OMITTED]
Table 2. Heat Pump Manufacturer's Rated Coefficient of Performance (COP) at Given Entering Water Temperature (EWT) Mode: Heating Cooling Stage: Low High Low Fan Speed: Low High Low High Low High Low EWT, [degrees] F ([degrees] COP COP COP COP COP COP COP C) House 1 (2.6 60 5.39 5.60 4.86 5.05 9.23 9.09 7.12 Ton) 70 5.97 6.25 5.15 5.39 8.09 7.97 6.33 80 6.29 6.68 5.28 5.60 6.71 6.59 5.51 90 6.65 7.15 5.39 5.80 5.39 5.30 4.75 House 2 (3.8 60 5.54 5.64 4.85 4.99 9.17 9.26 6.54 Ton) 70 6.04 6.15 5.05 5.25 8.00 8.06 5.92 80 6.46 6.58 5.17 5.43 6.74 6.80 5.19 90 6.87 7.01 5.26 5.58 5.48 5.54 4.51 High High COP House 1 (2.6 7.00 Ton) 6.21 5.42 4.66 House 2 (3.8 6.57 Ton) 5.95 5.25 4.57 Table 3. Seasonal Average System Coefficient of Performance (COP) House 1 COP House 2 COP Winter Early 2010 (heating) 4.86 2.65 Summer 2010 (cooling) 5.24 4.26 Winter 2010-2011 (heating) 3.44 2.36
The seasonal average COPs for the GSHP systems in the houses were lower than manufacturers' rated steady-state efficiencies for the same EWTs, with House 1 performing better than House 2. Measured 2010 cooling season COP was 5.24 for House 1, 3% to 35% below the manufacturer's range of 5.42 to 8.09 for the EWT experienced during the summer. Measured 2010 cooling season COP was 4.26 for House 2, 16% to 47% below the manufacturer's range of 5.09 to 8.06 for the EWT experienced during the summer.
The heating season COP was 4.86 for House 1 during the winter of early 2010, 4% to 22% lower than the manufacturer's range of 5.05 to 6.25 for the EWT experienced during the winter. Observed COP decreased to 3.44 during the winter season of 2010 to 2011. Although worse than anticipated, after accounting for differences in source energy, this observed system COP was comparable to the rated performance of a high efficiency (90+% AFUE) gas furnace. The cause of the year to year decline in COP is unclear and would require more detailed measurement to determine.
The heating season COP for House 2 was 2.65 in early 2010, declining to an average of 2.36 for the winter of 2010 to 2011. This is at least 50% lower than the manufacturer's listed COP values for the EWT experienced during the winters.
It is somewhat expected that the field measured system COP is lower than the manufacturer's predictions. Although the values in Table 2 incorporate the same energy inputs (e.g., fan, compressor, and ground loop pump) as measured in the field, the ANSI/AHRI/ASHRAE/ISO 13256-1 standard rating conditions, such as a fan working against zero static pressure drop and a ground loop pump working against zero head pressure loss, are not possible with installed systems. In both cases, the non-zero values that occur for installed GSHP systems result in increased energy consumption.
It is unclear why the heating season COPs varied substantially between the two houses; however, the GSHP system installed in the two-story house was oversized. The monthly average system COP and corresponding EWT for both houses are shown in Figure 2. The inside temperature of House 1 was 17.2 [degrees] C (63 [degrees] F) in 2010 for the months of January through March and in 2011 for January and February. The low temperature in the latter months was due to a compressor malfunction that prevented the system from heating the house. Based on observations from adjoining data in 2011, where the inside temperature was 20.6 [degrees] C (69 [degrees] F), this appeared to have no impact on the observed COP.
Although it was not measured at what stage or fan speed the system was operating, knowing this information would help determine if system oversizing was the cause of the difference in the COP between the two systems. The substantial difference between the rated and measured COP highlights the need for measurement and analysis of factors related to the quality of system design and installation, such as ground loop head pressure loss and air side external static pressure drop.
Domestic Water Heating Efficiency
Based on monitored data, the annual site energy EF of the domestic water heating system in House 1 ranged from 2.8 to 3.3, depending on the volume of water consumed. By comparison, electric resistance water heaters range in EF from 0.85 to 0.95. If compared to a natural gas water heater, the difference in source energy between the two fuels must be accounted for by converting the EF of each water heater to source energy weighted EF using Equations 6 and 7,
[EF.sub.SourceEnergyWeighted] = [EF.sub.ElectricDWH]/3.365 (6)
[EF.sub.SourceEnergyWeighted] = [EF.sub.GasDWH]/1.092 (7)
[EF.sub.ElectricDWH] is the EF for an electric DWH
[EF.sub.GasDWH] is the EF for a gas DWHP
3.365 is the site to source conversion for electricity (Deru and Torcellini 2007)
1.092 is the site to source conversion for natural gas (Deru and Torcellini 2007)
After accounting for the difference in source energy, Figure 3 shows that the DWH system in House 1 is on par with, or better than, the rated source energy efficiency of tankless gas water heaters and is more efficient than other options, including tank type gas, tank type electric, and heat pump domestic water heaters. In House 2, the desuperheater provided, on average, 20% of the thermal energy input to the tank, comparable to results obtained by Fanney and Dougherty (1992); however, there was no apparent improvement in EF over the rated performance of a conventional tank type electric water heater. More detailed measurements are necessary to determine if other factors, such as tank standby losses or disruption of tank stratification by the desuperheater, influenced the performance.
[FIGURE 3 OMITTED]
The results from this study indicate that GSHPs can be an effective means of providing space heating and cooling in warm climates; however, the variation in performance between the units in the two houses and when compared to manufacturers' data speaks to the potential impact of proper system sizing, design, and installation. Though no specific issues were documented with respect to the installation of the GSHP systems, apart from the oversizing of the system in House 2, more research of typical installed conditions is needed to understand the potential impact of loop design and layout on pressure head loss and air side ductwork on static pressure loss. It is important for homebuilders considering installing GSHP systems to understand that, due to the conditions experienced in the field, they should not expect installed system COPs to reach published unit COPs determined in bench tests.
In general, a desuperheater is a more efficient means of providing domestic hot water, but its efficiency varies, depending on the components to which it is connected. Based on these results, further research should be performed regarding the performance of desuperheaters in other residential applications, such as those connected to modern variable capacity air source heat pumps. As the desire to increase efficiency continues, integration of space conditioning and hot water generation systems, either directly (desuperheaters) or indirectly (heat pump water heaters), will create new challenges for calculating the in situ efficiency of operation of both devices.
The authors would like to thank the U.S. Department of Energy's Building America program for support to conduct this research, and Pine Mountain Builders, especially David Barnes and Marc Hoefert, for their assistance on site. Thanks also go to Bill Rittelmann and Chris Imm, who were vital to the test procedure development and instrumentation installation, respectively.
ASHRAE, 2005. ANSI/ARI/ASHRAE ISO 13256-1. Water-source heat pumps--Testing and rating for performance. Atlanta: American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
Deru, M. and Torcellini P. 2007, "Source Energy and Emission Factors for Energy Use in Buildings" Technical Report NREL/TP-550-38617, June.
Fanney, A. and B. Dougherty. 1992. Performance of a residential desuperheater. ASHRAE Winter Meeting, January 25-29, 1992. Anaheim, CA, pp. 489-499.
Hendron, R. 2008. NREL/TP -550-42662. Building America Research Benchmark Definition, Updated. December 20, 2007. Golden, CO: NREL. January.
Hendron, R. and C. Engebrecht. 2010. Technical Report NREL/TP-550-47246: Building America House Simulation Protocols. Golden, CO: NREL. January.
IBACOS. 2008. DE-FC26-08NT02231 Building America Program Annual Report BP 1. Pittsburgh: IBACOS.
IBACOS. 2010. DE-FC26-08NT02231 Building America Program Annual Report BP 2. Pittsburgh: IBACOS.
Lund, J., B. Sanner, L. Rybach, R. Curtis, and G. Hellstrom. 2004. Geothermal (Ground-Source) Heat Pumps--A World Overview. GHC Bulletin. September.
Merrigan, T. and D. Parker. 1990. FSEC-PF-215-90. Electrical Use, Efficiency, and Peak Demand of Electric Resistance, Heat Pump, Desuperheater, and Solar Hot Water Systems. Presented at American Council for an Energy Efficient Economy, August 1990, Asilomar Conference Center, Pacific Grove, CA.
Dave Stecher is a building performance specialist at IBACOS, Pittsburgh, PA. Kate Allison is a civil engineering undergraduate student at the University of Connecticut, Storrs, CT.
|Printer friendly Cite/link Email Feedback|
|Author:||Stecher, Dave; Allison, Kate|
|Date:||Jan 1, 2012|
|Previous Article:||Cold climates heat pump design optimization.|
|Next Article:||Large scale residential/municipal RES integrated refurbishment construction and HVAC systems engineering R&D needs.|