- Email: [email protected]
Measured effect of airflow and refrigerant charge on the seasonal performance of an air-source heat pump using R-410A
Energy and Buildings 43 (2011) 1802–1810
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Measured effect of airflow and refrigerant charge on the seasonal performance of an air-source heat pump using R-410A Larry Palmiter a , Jun-Hyeung Kim b,∗ , Ben Larson a , Paul W. Francisco c , Eckhard A. Groll d , James E. Braun d a
Ecotope, Inc., Seattle, WA 98105, USA School of Mechanical Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA c Building Research Council, University of Illinois, Urbana-Champaign, IL 61280, USA d School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USA b
a r t i c l e
i n f o
Article history: Received 7 December 2010 Received in revised form 2 February 2011 Accepted 22 March 2011 Keywords: Seasonal performance Heat pumps Refrigerant charge Airflow
a b s t r a c t The objective of this study was to measure the effects of improper airflow and refrigerant charge on the seasonal performance of a typical 10.6 kW, R-410A residential heat pump with a thermostatic expansion valve. Heating and cooling tests were performed in combinations of three refrigerant charges of 75%, 100%, and 125% of nominal value and two airflows of 75% and 100% of rated airflow. In addition, cyclic tests were performed to estimate the heating and cooling seasonal coefficient of performance (COP) at six climate zones specified by Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard 210/240-2008. Results showed that, in each climate zone, increases in refrigerant charge at the rated airflow could improve the unit’s heating seasonal COP by as much as 5%. However, combined decreases in airflow and refrigerant charge could penalize the unit’s heating seasonal COP by as much as 10%. © 2011 Elsevier B.V. All rights reserved.
1. Introduction and background Heat pumps have enjoyed a significant increase in popularity in recent years in the Pacific Northwest, both with the public and with utility program designers. Their popularity is due in part to their ability to save overall electricity throughout the heating season and reduce peak heating loads when compared to electric resistance heating. It is important to note that the savings will only be fully realized for properly installed and maintained equipment. In order to evaluate the benefits of heat pumps in utility programs, it is necessary to understand the degree to which heat pump performance is affected by incorrect charge and airflow and to focus efforts on ensuring that appropriate levels of effort are taken to avoid substantial penalties [1]. Previous studies have demonstrated that incorrect refrigerant charge and improper airflow would cause the performance of air
Abbreviations: Cd , coefficient of degradation; COP, coefficient of performance; IDDB, indoor dry bulb temperature; HSPF, heating seasonal performance factor; ODDB, outdoor dry bulb temperature; ODWB, outdoor wet bulb temperature; PLF, part load factor; SEER, seasonal energy efficiency ratio; TXV, thermostatic expansion valve. ∗ Corresponding author. Tel.: +1 205 348 4492; fax: +1 205 348 6419. E-mail addresses: [email protected] (L. Palmiter), [email protected] (J.-H. Kim), [email protected] (B. Larson), [email protected] (P.W. Francisco), [email protected] (E.A. Groll), [email protected] (J.E. Braun). 0378-7788/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2011.03.026
conditioners to degrade. Mowris et al. [2] reported that about 6 out of 10 air conditioners in the field were operated with improper refrigerant charge and airflow and that this caused the air conditioners to underperform by about 10–20%. Their study strongly suggests that heat pumps in field operation would have a similar problem. Domingorena and Ball [3] measured the effect of refrigerant charge varying from 15% below to 25% above the nominal value on heat pump heating performance. The heating performance remained relatively unchanged at both undercharged and overcharged conditions. They attributed this performance insensitivity to the suction-line accumulator. O’Neal and Farzad [4] studied the effect of improper refrigerant charge on the seasonal cooling performance of a 10.6 kW R-22 air conditioner with capillary tube expansion. They carried out steadystate and cycling cooling performance tests while changing the refrigerant charge from 20% below to 20% above the full charge. The cooling seasonal COP peaked at the full charge of 4.0 kg. However, at 20% undercharged and 20% overcharged conditions, the cooling seasonal COP decreased by 20% and 10%, respectively. Shen et al. [5] measured the cooling performance of a 10.6 kW R410A unitary split heat pump with a thermostatic expansion valve (TXV) under reduced evaporator airflow at dry and wet conditions. Results showed that the cooling performance was degraded much more at dry conditions than at wet conditions. A 20% reduction in evaporator airflow at dry conditions caused 14% and 10% drops in
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
1803
Table 1 Heating mode testing conditions. Test description
Air entering indoor unit temperature Dry-bulb (◦ C)
H1 test (steady) H1C test (cyclic) H3 test (steady) 35 test (steady)a a
21.1 21.1 21.1 21.1
Air entering outdoor unit temperature Wet-bulb (◦ C) max
15.6 15.6max 15.6max 15.6max
Dry-bulb (◦ C) 8.3 8.3 −8.3 1.7
Wet-bulb (◦ C) 6.1 6.1 −9.4 –
Test results at the “35” conditions were used to estimate the defrost penalty at the H2 (frost accumulation) conditions.
cooling capacity and COP, respectively, while the same reduction in evaporator airflow at wet conditions resulted in 5% drops in both cooling capacity and COP. Results also indicated that the cooling performance at wet conditions was relatively unchanged until the evaporator airflow decreased 25%. These studies indicate that improper refrigerant charge and incorrect blower airflow would penalize the performance of heat pumps in both heating and cooling mode; however, their results are mostly limited to steady-state operation. No cycling tests have been performed with heat pumps in heating mode, which is essential for studying the effect of improper refrigerant charge and blower airflow on the heating seasonal COP of heat pumps. Furthermore, no study has been found in the literature examining the combined effect of improper refrigerant charge and blower airflow on the heating seasonal COP of heat pumps. The seasonal performance of heat pumps is measured in terms of heating and cooling seasonal COP values,1 which are set forth by test procedures in the Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard. 210/240 [6]. Since the actual heating performance of heat pumps changes with different climate conditions, AHRI Standard 210/240 [6] specifies six different climate conditions, i.e., Climate Zones 1 through 6. However, only the rated heating seasonal COP of heat pumps, determined based on Climate Zone 4, are normally made available to the public, while the other heating seasonal COP values are not reported. Although the rated heating seasonal COP does not represent the predicted performance in other climate zones other than Climate Zone 4, it is commonly used to estimate electricity use during the heating season in different climate zones. For instance, for Climate Zone 1, which covers most of the area of the state of Florida in the U.S., the use of a rated heating seasonal COP would overestimate the actual electricity use during the heating season. Thus, in order to allow for climate-related variations in heat pump performance, it is necessary to determine the heating seasonal COP values of heat pumps in all six climate zones and to use the proper heating seasonal COP value for each climate zone in estimating the annual heating costs. The goal of this study is to determine the impact of refrigerant charge and blower airflow variations on the seasonal performance of a 10.6 kW R-410A residential heat pump. This study does not specifically attempt to determine the rated performance of the unit, but rather to measure the expected in situ performance which could result from a variety of typical installation practices [1]. All tests were performed by closely following the methods used in the rating standards to establish a common base for performance comparisons. All test procedures and results for the heat pump are summarized in this paper, and the effects of improper refrigerant charge and blower airflow on the seasonal heating and cooling performance are then discussed for all six climate zones in the U.S.
1 Heating and cooling seasonal COP values are determined by dividing the heating seasonal performance factor (HSPF) and seasonal energy efficiency ratio (SEER) each by 3.412 Btu/Wh.
2. Test method The medium-performance heat pump tested was a 10.6 kW R-410A model with a scroll compressor and a suction-line accumulator. The outdoor unit had a one-row, 787-fins-per-meter coil with a face area of 1.87 m2 . The compressor was rated at 79 locked rotor amps and 17.1 rated load amps. The outdoor fan was a propellertype, direct-drive fan rated at 0.9 full load amps. The indoor unit had a three-row, 570.9-fins-per-meter coil with a face area of 0.32 m2 . The indoor fan was a centrifugal-type, direct-drive fan. The heat pump included both a time-delay relay and a TXV with a charging subcooling of 5.6 K. As the project was to measure and compare the system performance of a new, factory-provided heat pump at different charge levels and airflow rates, the unit was purchased from a local distributor and installed according to the installation manual with minimum intervention. The only exception to a completely standard setup process was the charge adjustment to properly account for the installed liquid refrigerant line length per the installation instructions. All laboratory tests were done to characterize the performance at the rating conditions specified in Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard 210/240 [6]. A set of tests at extremely hot conditions such as those found in the Southwest U.S. desert was also done. The tests were conducted at each of three outdoor temperatures in heating mode (−8.3 ◦ C, 1.7 ◦ C, and 8.3 ◦ C) and in cooling mode (27.8 ◦ C, 35 ◦ C, and 51.7 ◦ C). These tests provide useful comparison to the manufacturer’s published heating and cooling seasonal COP values for the unit. Additionally, in order to measure off-design performance in both modes, refrigerant charges were varied from approximately 25% above to 25% below the manufacturer’s nominal recommended charge at each 75% and 100% of rated airflow. The airflow was varied using a booster fan in the laboratory, and the airflow adjustments in this study were specifically designed to target the effects resulting from a wide range of different conditions leading to different static pressures. In practice, many heat pump applications have flow rates lower than the nominal airflow rate, which can lead to problems with both cooling and heating performance. In heating mode, capacity, power, fan power and COP were measured in all combinations of the three outdoor temperatures of −8.3 ◦ C, 1.7 ◦ C, and 8.3 ◦ C, three refrigerant charges of 75%, 100%, and 125% of the nominal value, and two airflows of 0.4248 and 0.566 m3 /s (corresponding to 75% and 100% of rated airflow). The detailed heating mode test conditions are listed in Table 1. Partload performance tests were also conducted to characterize the coefficient of degradation (Cd ), which is essential for calculating the heating seasonal COP. The Cd measurement was made only at a temperature of 8.3 ◦ C (H1 steady and H1C cyclic test conditions). Cooling mode tests were performed at the cooling mode test conditions listed in Table 2. Understanding the cooling mode tests requires a more detailed discussion of refrigerant charge levels. The catalog-recommended charge is 3.36 kg of R-410A. The test setup had an extra 3.05 m of refrigerant line. Per manufacturer’s
1804
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Table 2 Cooling mode testing conditions. Test description
A test (steady, wet coil) B test (steady, wet coil) C test (steady, dry coil) D test (cyclic, dry coil) 125 testa (steady, dry coil) 120 testb (steady, dry coil) a b
Air entering indoor unit temperature
Air entering outdoor unit temperature
Dry-bulb (◦ C)
Wet-bulb (◦ C)
Dry-bulb (◦ C)
Wet-bulb (◦ C)
26.7 26.7 26.7 26.7 26.7 26.7
19.4 19.4 – – – –
35 27.8 27.8 27.8 51.7 48.9
– – – –
The 125 test conditions is added for hot desert test conditions in the Southwest U.S. For 117% charge, the 120 test conditions were used in replacement of the 125 test conditions due to high-side pressure switching.
Table 3 Performance comparison of laboratory tests with catalog data at 3.54 kg charge level and 0.5663 m3 /s. Test
B A 125 a b
Testing results/catalog dataa
Testing/catalog ratio data
Capacity (kW)
Power (kW)
COP
Cooling seasonal COPb
Capacity (%)
Power (%)
COP (%)
Cooling seasonal COP (%)
10.43/10.62 9.89/9.91 7.14/7.99
2.95/2.78 3.37/3.17 4.95/4.29
3.54/3.82 2.93/3.12 1.44/1.86
3.47/3.81
98.2 99.8 89.3
106.1 106.3 115.4
92.6 93.9 77.4
91.0
Capacity and power interpolated from table for the B test. Cooling seasonal COP calculated using a coefficient of degradation, Cd = 0.04, and a part load factor, PLF = 0.98.
recommendations, the test laboratory added 0.19 kg of refrigerant to compensate for the extra length resulting in a base charge (100%) of 3.54 kg. The high charge (125%) was then 4.43 kg and the low charge (75%) was 2.66 kg. All of the heating mode tests were done with these values, and cooling mode tests were done at 100% and 75% charge. It was planned to perform an additional set of tests at an outdoor temperature of 51.7 ◦ C (125 test condition) with dry indoor coil to examine the performance under hot, dry conditions. Unfortunately, at the 125% charge level, the high-pressure limit switched the compressor off at outdoor temperatures above 29.4 ◦ C. As a result, it was decided to run the high-charge tests at only 3.94 kg of charge, about 111% of the 3.54 kg base level. In the first base charge tests, it was noticed that the test results did not compare well with the catalog data. Table 3 shows the comparison with catalog data for the first base charge tests. In comparison to the catalog data, the test results show lower capacity and higher power resulting in a lower COP, especially at 51.7 ◦ C (125 test conditions), where the COP was 23% lower. The seasonal cooling COP was 9% below the catalog rating. In addition, the measured subcooling was 8.4 K, which is 2.8 K higher than the rated subcooling of 5.6 K. It was suggested that the addition of the extra 0.19 kg of refrigerant might have caused the relatively large difference between the experimental data and the catalog data. All of the base charge tests were redone at a charge of 3.36 kg, where the measured and rated subcoolings were matched. Table 4 compares these tests with the catalog data. The agreement is much better, especially at 51.7 ◦ C. The power comparisons are now very close. The cooling seasonal COP is now about 6.5% below the catalog rating. This unit is very sensitive to additional charge, even in a situation where the length of line set would jus-
tify the extra charge. The recommended charge is close to the upper limit that allows heat pump cooling operation at 51.7 ◦ C outdoor temperature. This may indicate the need for precautions to avoid overcharging this unit in hot climates. Due to the matched subcooling and better performance agreement, it was decided to use 3.36 kg as the base level (100%) for the cooling tests. The charge level of 2.66 kg was then at 79% of the base and the charge level of 3.94 kg was 117% of the base. Also, the hightemperature tests were done at an outdoor temperature of 48.9 ◦ C instead of 51.7 ◦ C for the high charge tests. As a result, cooling mode tests were performed for all six combinations of refrigerant charges of 79%, 100%, and 117% of the nominal value (3.36 kg of R-410A), and air flows of 0.4248 and 0.5663 m3 /s (corresponding to 75% and 100% of rated airflow). For each of the six combinations, steady-state tests were done at three different outdoor temperatures (27.8 ◦ C, 35 ◦ C, and 51.7 ◦ C, except for the high charge case where the outdoor temperature was 48.9 ◦ C). In addition, cyclic tests were carried out at an outdoor temperature of 27.8 ◦ C with a dry indoor coil to determine the Cd and subsequently the cooling seasonal COP. The heat pump tests were designed in accordance with AHRI Standard 210/240 [6]. For each combination of airflow, refrigerant charge, and outdoor temperature, a single test of the equipment was conducted. With the steady-state tests, the environmental chambers and the unit were operated until they reached the steady state conditions, and then the test data were recorded. For the cyclic tests, the unit was cycled on and off at least two times, and then immediately following the cycling, the test data were recorded. A number of measurements were recorded at each testing point including airflow, dew point, unit power consumption, air and
Table 4 Performance comparison of laboratory tests with catalog data at 3.36 kg charge level and 0.5663 m3 /s. Test
B A 125 a b
Testing results/catalog dataa
Testing/catalog ratio data
Capacity (kW)
Power (kW)
COP
Cooling seasonal COPb
Capacity (%)
Power (%)
COP (%)
Cooling seasonal COP (%)
10.50/10.62 9.67/9.91 7.34/7.99
2.83/2.78 3.20/3.17 4.42/4.29
3.71/3.82 3.02/3.12 1.66/1.86
3.56/3.81
98.8 97.6 91.8
101.8 100.9 103.0
97.1 96.7 89.1
93.5
Capacity and power interpolated from table for the B test. Cooling seasonal COP calculated using a coefficient of degradation, Cd = 0.08, and a part load factor, PLF = 0.96.
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
1805
Table 5 Normalization values of heating capacity and COP at nominal charge and airflow. Test condition
Capacity (kW)
COP
H1 (Tout = 8.3 ◦ C) 35 (Tout = 1.7 ◦ C) H3 (Tout = −8.3 ◦ C)
10.71 8.87 6.24
3.51 3.08 2.31
3.1. Heating capacity
Fig. 1. Heating capacity ratio versus airflow rate at the H3, 35, and H1 test conditions.
refrigerant temperatures and pressures, and refrigerant mass flow rate. • Airflow was measured with a nozzle apparatus, built in accordance with ASHRAE standard 116 [7] and having an uncertainty of ±10%. • The hygrometer to measure dew point was accurate within ±0.2 ◦ C. • For temperature measurements in both the air and the refrigerant, T-type thermocouples with an error of ±1 ◦ C were employed. For the air-side measurements, eight thermocouples were arranged in a grid placed in the flow path and averaged to determine the temperature. • Seven pressure transducers were installed to measure pressure on the refrigerant side. Three had an error of ±0.13% of full scale and four had an error of ±1.0% of full scale. • A Coriolis flow meter, with an error of ±0.5% of full scale, was installed in the liquid line to measure refrigerant mass flow rate. • To measure indoor blower and outdoor fan power consumptions and compressor power consumption, digital watt transducers with an accuracy of ±0.2% were used. 3. Heating mode test results This section summarizes the effects of refrigerant charge and airflow on capacity, COP, Cd , and the defrost penalty. Figs. 1 and 2 show the capacity and COP ratios, respectively, at the H3, 35, and H1 test conditions. Under each test conditions, capacities and COP values are normalized to those measured at 100% refrigerant charge (3.54 kg) and 0.5663 m3 /s airflow (100% of rated airflow). Table 5 shows the capacities and COP values used to normalize each point.
Fig. 1 shows the effect of airflow on capacity for each of the nine combined cases of test conditions and charge level (e.g., in the case of H3-75% charge, H3 indicates test conditions, and 75% charge shows charge level). In all nine cases, the capacity decreases about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). Between the nine cases at 0.5663 m3 /s (100% of rated airflow), the 75% charge level reduces capacity by approximately 3% and 8% at the 35 (1.7 ◦ C) and H1 (8.3 ◦ C) test conditions, respectively, compared to their 100% charge levels. However, at the H3 (−8.3 ◦ C) test conditions, the 75% charge level actually increases capacity by 5%. On the other hand, the 125% charge level does not affect capacity at either the 35 (1.7 ◦ C) or H1 (8.3 ◦ C) test conditions, but improves capacity by 3% at the H3 (−8.3◦ C) test conditions. 3.2. Coefficient of performance for heating The coefficient of performance (COP) is also calculated using AHRI Standard 210/240 [6], which includes the indoor blower and outdoor fan power as well as that of the compressor in the denominator. The heat generated by the indoor fan is included in the capacity in the numerator. Each of these values was measured in the laboratory. AHRI mandates that for test units that do not include a specific indoor air handler, a default value for the indoor blower power of 365 W per 0.4719 m3 /s must be used. The indoor fan power measurements for the model tested averaged about 391 W at 0.4248 m3 /s (75% of rated airflow) and 533 W at 0.5663 m3 /s (100% of rated airflow). The outdoor fan measurements for the current tests averaged 212 W. Fig. 2 shows the effect of airflow on heating COP for each of the nine combined cases of test conditions and charge level (e.g., in the case of H3-75% charge, H3 indicates the test conditions, and 75% charge shows the charge level). In all nine cases, the COP decreases by about 6% for a 25% reduction in airflow. The effect of charge on COP is within about ±3%. Figs. 1 and 2 clearly show that the unit should not be run with airflow below the nominal airflow due to the performance penalties. 3.3. Part-load and defrost penalty factors
Fig. 2. Heating COP ratio versus airflow rate at the H3, 35, and H1 test conditions.
Two other factors have a large effect on heat pump performance: part-load operation and the defrost cycle. Under part-load conditions, the heat pump cycles off and on. For a short period during the start-up the heat pump draws nearly full power, but there is no output while the appropriate equilibrium conditions are being established throughout the refrigerant side. The indoor fan is off during this period. Additionally, each time the unit cycles off there is heat loss in the system. The net effect of these losses is an increasing loss of efficiency as the unit runs for a smaller fraction of time. The coefficient of degradation (Cd ) is the efficiency loss that occurs due to the cycling of the unit. AHRI Standard 210/240 [6] assumes a linear percentage loss between zero load and full load (where the efficiency reaches the steady-state value). For instance, with a Cd of 0.25 the efficiency at zero load is reduced to 75% of the steady-
1806
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Fig. 3. Cd versus airflow rate.
state value, and at 50% load the efficiency is reduced to 87.5% of the steady-state value. An additional performance loss occurs under outdoor conditions that lead to ice buildup on the outdoor coil. During the defrost cycle the outdoor fan is off, and the heat pump operates as an air-conditioner to warm the outdoor coils. In order to avoid occupant discomfort, typically the indoor fan and backup heat also run during this cycle to warm the air coming off the cold indoor coil. Unfortunately, due to contract deadlines, the defrost tests at the H2 conditions (1.7 ◦ C ODDB, 0.6 ◦ C ODWB, and 21.1 ◦ C IDDB) were not conducted. However, the defrost penalty can be stated as a multiplier of the steady-state efficiency measured at the 35 test conditions [8]. A multiplier of 0.9 [9] was used to estimate the penalized heating capacity and COP at the H2 conditions for the test unit, and the estimated heating capacity and COP were used for the heating seasonal COP calculations. The AHRI Standard requires the measurement of Cd and the defrost penalty, but neither heat pump manufacturers nor AHRI publish these measured values. Although unpublished, they are used by the manufacturer to calculate the heating seasonal COP. In this study, one of the goals is to perform all of the required measurements needed to calculate the heating seasonal COP for various combinations of airflow and charge level. Fig. 3 illustrates the effect of airflow and charge on the Cd . It shows that the impact of airflow on the Cd is only a few percent at each charge level, and the Cd at 75% charge is about 0.3, at nominal charge about 0.25 and at 125% charge about 0.16. However, the impact of charge on the Cd is significant. The Cd at 75% charge is about 20% higher than that at nominal charge while the Cd at 125% charge is about 36% lower. This suggests that the heat pump can produce heating at each start-up more quickly when it is over-charged to some extent. 4. Heating seasonal performance It is difficult to estimate the seasonal performance of a heat pump because both the capacity and the COP depend strongly on outdoor temperature. In addition, the capacity at low outdoor temperatures is usually not adequate to meet the heating load, thus requiring the use of backup heat. Also, the effects of part-load operation and defrost must be taken into account. In the late 1970s a bin-hour calculation method that accounts, to some extent, for all of these effects was developed. This method results in the heating seasonal COP.2 According to AHRI Standard 210/240 [6], Section 4.2, heat pump manufacturers are required to calculate the heating seasonal COP for given bin-hour profiles for six different climate
2 Heating seasonal COP is determined by dividing the heating seasonal performance factor (HSPF) by 3.412 Btu/Wh.
zones “unless an approved alternative rating method is used, as set forth in 10 CFR 430.24(m), Subpart B [10].” However, the U.S. government only allows the heating seasonal COP for Climate Zone 4 to appear on the label. This is often misleading for performance in other regions because the heating seasonal COP varies strongly with climate zone and heating load [11]. Climate and heating load differences also make it so that the relative benefits of a more efficient heat pump compared to a less efficient heat pump are not consistent between climate zones [12]. For example, a heat pump with a 10% higher published (Climate Zone 4) heating seasonal COP than another will not result in a 10% efficiency improvement in all climates, with some climates having lesser benefits and some climates having greater benefits. The laboratory test data were used in this study for calculating the heating seasonal COP in accordance with AHRI Standard 210/240 [6] for all six climate zones. 4.1. Calculation method AHRI Standard 210/240 [6] defines the test methods used to measure the heating seasonal COP of a heat pump as well as the equations required for calculation. The necessary variables for the calculation are airflow rates, heating capacities and electrical power consumptions from steady-state tests at −8.3 ◦ C (H3 test conditions), 1.7 ◦ C (35 test conditions), and 8.3 ◦ C (H1 test conditions), a cyclic test at 8.3 ◦ C, and a defrost test at 1.7 ◦ C. The capacity and input power at other outdoor temperatures are estimated as follows. In the capacity and input power versus outdoor temperature plots, the slope of the line below −8.3 ◦ C and above 7.2 ◦ C is equal to that of a line connecting the 1.7 ◦ C and 8.3 ◦ C test point values. The central portions of the curves are defined by connecting the −8.3 ◦ C and 1.7 ◦ C test values and the newly defined point at 7.2 ◦ C. The load, capacity, compressor input, and auxiliary heat (assumed all-electric) are then calculated for each temperature bin, applying part-load corrections as needed. There are some assumptions implicit in this method. It assumes that there is no defrost penalty below −8.3 ◦ C; however, for units with time–temperature defrost control there will be significant defrost penalties at all temperatures below about 4.4 ◦ C. There is also a large discontinuity in the performance curves at 7.2 ◦ C, an effect that does not occur in the laboratory. Additionally, the house load assumes a home heating balance point of 18.3 ◦ C (meaning the mimimum outdoor temperature at which no heat is required by the home), which is on the high side for well-insulated homes. This is compensated somewhat by multiplication of the load at each outdoor temperature by a load factor of 0.77. The load factor procedure, however, is no longer sanctioned by ASHRAE. ASHRAE recognized the problem of using the load factor as early as the 1985 Handbook of Fundamentals [13] and by the 1993 Handbook of Fundamentals [14] the load factor was no longer included in the discussion of energy estimating methods. Instead the variable-base degree day method was used. Nevertheless, heating seasonal COP serves as a single point rating for heat pump performance that attempts to account for all of the major factors affecting performance. The largest problem lies in failing to publish the ratings for all six zones (although at least one major manufacturer publishes heating seasonal COP ratings for Climate Zone 5 in addition to Climate Zone 4). 4.2. Heating seasonal COP bin data Fig. 4 shows the fractional temperature bin data from AHRI Standard 210/240 [6] used for the six climate zones. The zones pertinent to the Pacific Northwest are 4, 5, and 6. The climates of Boise, ID and Spokane, WA are well represented by Climate Zone 4 (the rating Climate Zone), while Missoula, MT is in Climate Zone 5. Climate Zone 6 is representative of the climate west of the Cascade Mountains in
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Fig. 4. Fractional bin hour data for heating seasonal COP calculations (AHRI Standard 210/240 [6]). Table 6 Heating seasonal COP at nominal charge and airflow. Zone
1
2
3
4
5
6
Heating seasonal COP
2.69
2.63
2.50
2.29
1.95
2.64
Oregon and Washington, e.g., Seattle, WA and Portland, OR. It has a peak in the relative bin temperature distribution at about 10 ◦ C and very few hours at cold temperatures. It should be noted that the Climate Zone map of the US provided in the AHRI Standard 210/240 [6] was incorrectly drawn in the first publication of the Standard and, unfortunately, has never been corrected. This error results in the assignment of completely incorrect climate zones for the region west of the Cascade Mountains in Oregon, Washington, and Northern California. For example, Seattle lies on the border between Climate Zones 4 and 5 and Portland is fully in Climate Zone 5 in the map. This area west of the Cascades is actually Climate Zone 6, which has a level of heat pump performance in heating mode comparable with that in Climate Zone 1.
1807
Fig. 6. Normalized heating seasonal COP versus airflow rate in zones 4, 5, and 6.
Figs. 5 and 6 show the effect of airflow and charge on heating seasonal COP for each of the six climate zones. The heating seasonal COP decreases by about 3–4% when the airflow is reduced from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for each climate zone and charge level. Reducing the charge to 75% results in an 8% loss in heating seasonal COP and increasing the charge to 125% results in an 8% gain for the warmer climate zones (1, 2, 3 and 6). For the colder climates, the effect of a 25% change in charge is about 3–5%. 5. Cooling mode results
Table 6 shows heating seasonal COP values calculated from the laboratory test data for nominal charge and airflow. These values were used to normalize the heating seasonal COP plots in Figs. 5 and 6. It should be noted that there is a large variation in heating seasonal COP across climate zones ranging from the lowest value of 1.95 in Zone 5 to the largest value of 2.69 in Zone 1. The measured value of 2.29 in Zone 4 shows excellent agreement with the published heating seasonal COP for this heat pump of 2.32.
In cooling mode, the standard tests required by AHRI Standard 210/240 [6] were performed. These tests were made at outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions) with wet indoor coil (entering wet bulb temperature of 19.4 ◦ C and dry bulb of 26.7 ◦ C). In order to normalize the performance graphs, the measured capacity and COP at 0.5663 m3 /s (100% of rated airflow) and at 3.36 kg of charge (100% charge) in Table 4 were used. As expected, the capacity and COP decrease rapidly with increasing outdoor temperature. Fig. 7 shows the normalized cooling capacity ratios as a function of airflow and charge level at each outdoor temperature. The capacity decreases by about 6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions). For the highest outdoor temperatures, the airflow effect was about 10%. At 79% charge, the capacity is reduced 10% or more for the lower two temperatures across all airflow levels. For the highest outdoor temperatures at 79% charge the capacity is reduced over 10% for 0.4248 m3 /s airflow (75% of rated airflow).
Fig. 5. Normalized heating seasonal COP versus airflow rate in zones 1, 2, and 3.
Fig. 7. Cooling capacity ratio versus airflow rate at the B, A, and 125 (or 120) test conditions.
4.3. Effect of charge and airflow on heating seasonal COP
1808
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
6. Conclusions
Fig. 8. Cooling COP ratio versus airflow rate at the B, A, and 125 (or 120) test conditions.
Results show that the heating seasonal COP of the heat pump differs widely between the six climate zones in the U.S, ranging from 1.95 in Zone 5 to 2.69 in Zone 1, while it is 2.29 in Zone 4. Results also show that, in each climate zone, different airflow and refrigerant charge for the heat pump would make a noticeable impact on the unit’s seasonal performance (i.e., heating and cooling COP). For instance, increased charge tends to have substantial benefits in heating mode. However, the sharp performance drop in cooling mode, including shutting down due to the triggering of the high pressure limit switch at high temperatures, indicates that caution must be taken with regard to overcharging a unit if cooling is of substantial interest. Decreased airflow tends to produce significant seasonal performance penalties in both heating and cooling mode. This shows the importance of proper airflow during the heating and cooling seasons. 6.1. Heating mode The major findings of the heating mode laboratory tests and the heating seasonal COP calculations are summarized below.
Fig. 9. Cooling seasonal COP ratio versus airflow rate.
Fig. 8 shows the normalized cooling COP ratios at two different airflows. At 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions), the COP decreases by about 2% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). At high temperatures of 48.9 ◦ C (120 test conditions) and 51.7 ◦ C (125 test conditions), the COP is penalized by 4–10% as airflow decreases. Fig. 8 also shows that overcharging or undercharging the unit results in COP decreases of at least 8% for these temperatures at each airflow. Figs. 9 and 10 show the impact of charge and airflow on the cooling seasonal COP and the coefficient of degradation Cd , respectively. At each refrigerant charge, changes in airflow do not make a noticeable impact on the cooling seasonal COP. However, increasing or reducing charge changes the cooling seasonal COP with reductions of 3–4% for high-charge and 10% for low-charge at each airflow rate. The Cd demonstrates a strong dependence on charge. At nominal airflow it ranges from near 0.00 at 117% charge to 0.09 at 79%. The effect of airflow can be seen at 79 and 100% charge with Cd ranging from 0.07 to 0.09. Airflow has no significant effect at 117% charge.
Fig. 10. Coefficient of degradation versus airflow rate.
• Heat pump capacity decreases about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). The most significant impact of reduced charge (8–10%) is at 8.3 ◦ C where the loss in capacity will have little impact due to small heating loads. • Heat pump COP shows decreases of about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) at each charge level and temperature. In contrast, the impact of charge on COP is quite small (1–3%). • The coefficient of degradation, Cd , at 75% charge is about 0.3 and at 100% charge about 0.25. The impact of airflow is only a few percent. • The calculated heating seasonal COP at nominal conditions varies noticeably across climate zones from 1.95 in Zone 5 to 2.69 in Zone 1. The value of 2.29 calculated for Zone 4 shows excellent agreement with the published value for this heat pump of 2.32. • The heating seasonal COP decreases by about 3 to 4% when the air flow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for each climate zone and charge level. Reducing the charge to 75% results in an 8% loss in heating seasonal COP, and increasing the charge to 125% results in an 8% gain for the warmer climate zones (1, 2, 3 and 6). For the colder climates, the effect of a 25% change in charge is about 3–5%. 6.2. Cooling mode The major findings for cooling mode are summarized below. In cooling mode the performance is more strongly affected by both airflow and charge. • Difficulties in operating the equipment were encountered at high temperatures with even moderate levels of overcharge. This stresses the need for precautions to prevent overcharging, especially when the unit is used in hot climates. • The capacity decreases by about 6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions). For the highest outdoor temperatures, the airflow effect is about 10%. • At 79% charge, the capacity is reduced 10% or more for the lower two temperatures across all airflow levels. For the highest out-
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
•
•
•
•
door temperatures at 79% charge, the capacity is reduced over 10% for 0.4248 m3 /s airflow (75% of rated airflow). At 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions), COP decreases by about 2% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). At high temperatures of 48.9 ◦ C (120 test conditions) and 51.7 ◦ C (125 test conditions), the COP is penalized by 4–10% as the airflow decreases. Overcharging or undercharging the unit results in a COP decrease of at least 8% for these temperatures. The effect of overcharging the unit significantly reduces the COP. At high charge, the COP decreases by at least 10% compared with that at nominal charge for all airflows. The cooling seasonal COP value shows a large dependence on charge level but little on airflow. 117% charge reduces the cooling seasonal COP by 3–4% while 79% charge reduces by 10%. The Cd also demonstrates a strong dependence on charge. At nominal airflow, it ranges from near 0 at 117% charge to 0.09 at 79% and 100% charge, which is less than the default Cd value of 0.25. Airflow has little effect on the Cd at all charge levels. The coefficient of degradation, Cd , at 79% and 100% charge ranges from 0.07 to 0.09. The Cd at 117% charge is zero.
1809
pump performance in general. In particular, a heat pump without a suction-line accumulator may be much more sensitive to variations in charge level. However, investigations of the impact of the accumulator relative to its absence were not within the scope of this study. These results only apply to the heat pump tested and should not be generalized across other heat pump models, even those using the same refrigerant. Additional testing is needed to identify the range of results possible across all available heat pumps. However, the results are examples of conditions that can be found in at least one heat pump model.
Acknowledgements This research was funded by the US Department of Energy, the Bonneville Power Administration, and the States of Oregon, Washington, Idaho, and Montana. The project manager was Ken Eklund of the Idaho Energy Division.
Appendix A. 7. Limitations and suggestions The results suggest that caution should be taken in making any statements about the effects of charge and airflow on heat
Performance test data for the 10.6 kW heat pump including cooling and heating capacities, compressor powers, fan powers, and COPs are tabulated (Tables A.1–A.4).
Table A.1 Cooling mode test data at steady-state conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
A A B B C C 125 125
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
2.66 2.66 2.66 2.66 2.66 2.66 2.66 2.66
A A B B C C 125 125
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
A A B B C C 120 120
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
Indoor net cooling (kW)
Indoor sensible cooling (kW)
COP (–)
Indoor fan power (W)
Outdoor fan power (W)
Outdoor compressor power (W)
8.29 8.79 8.81 9.36 7.97 8.78 6.26 6.71
6.12 7.13 6.39 7.35 7.97 8.78 6.26 6.71
2.77 2.81 3.33 3.37 3.00 3.16 1.52 1.58
396 523 397 523 403 526 400 529
196 195 198 198 201 197 195 195
2405 2405 2051 2058 2049 2051 3529 3534
3.36 3.36 3.36 3.36 3.36 3.36 3.36 3.36
9.08 9.67 9.78 10.50 8.57 9.52 6.53 7.34
6.53 7.56 6.83 7.95 8.57 9.52 6.53 7.34
2.97 3.02 3.62 3.71 3.21 3.40 1.51 1.66
384 513 387 510 386 510 382 509
191 193 195 195 197 196 187 187
2484 2490 2118 2123 2089 2095 3744 3726
3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94
9.17 9.77 9.83 10.52 8.42 9.53 6.59 7.39
6.62 7.66 6.87 7.96 8.42 9.53 6.59 7.39
2.70 2.78 3.34 3.44 2.87 3.10 1.33 1.47
384 510 386 510 390 519 384 510
192 192 195 195 196 196 190 189
2818 2815 2361 2356 2350 2357 4388 4324
Table A.2 Cooling mode test data at cyclic conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
Cyclic indoor ave. net cooling (kW)
Cyclic indoor ave. fan power (W)
Cyclic outdoor ave. fan power (W)
Cyclic outdoor ave. comp power (W)
Cyclic average COP (–)
Cyclic coefficient degradation (–)
D D
0.4248 0.5660
2.66 2.66
1.57 1.72
104 136
40 39
410.2 411.4
2.83 2.93
0.07 0.09
D D
0.4248 0.5660
3.36 3.36
1.66 1.84
101 128
39 39
408.0 413.7
3.03 3.17
0.07 0.08
D D
0.4248 0.5660
3.94 3.94
1.64 1.84
102 133
39 39
411.0 415.4
2.98 3.14
0 0
1810
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Table A.3 Heating mode test data at steady-state conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
2.66 2.66 2.66 2.66 2.66 2.66
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
Indoor coil net heating (kW)
COP (–)
Indoor fan power (W)
Outdoor fan power (W)
Outdoor compressor power (W)
9.50 9.91 6.08 6.58 8.21 8.63
3.19 3.38 2.25 2.40 2.86 3.02
387 529 394 538 392 532
209 210 216 218 211 211
2385 2189 2088 1991 2272 2112
3.54 3.54 3.54 3.54 3.54 3.54
10.06 10.71 6.00 6.24 8.23 8.87
3.23 3.51 2.25 2.31 2.86 3.08
389 533 393 531 386 531
209 210 215 216 210 211
2519 2311 2063 1951 2285 2137
4.43 4.43 4.43 4.43 4.43 4.43
9.98 10.76 6.10 6.45 8.42 8.83
3.21 3.48 2.27 2.37 2.89 3.05
389 528 398 538 391 538
210 211 217 219 211 214
2506 2349 2067 1964 2307 2146
Table A.4 Heating mode test data at cyclic conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
Cyclic indoor ave. net heating (kW)
Cyclic indoor ave. fan power (W)
Cyclic outdoor ave. fan power (W)
Cyclic outdoor ave. comp power (W)
Cyclic average COP (–)
Cyclic coefficient degradation (–)
H1C H1C
0.4248 0.5663
2.66 2.66
1.38 1.45
104 138
42 42
427.6 403.8
2.40 2.49
0.29 0.31
H1C H1C
0.4248 0.5663
3.54 3.54
1.55 1.65
105 142
42 42
450.3 419.2
2.60 2.73
0.23 0.26
H1C H1C
0.4248 0.5663
4.43 4.43
1.74 1.88
102 137
42 42
471.2 450.5
2.83 2.98
0.14 0.18
References [1] B. Davis, D. Baylon, S. Strand, It sounded good on paper: field performance of residential heat pumps in the Pacific Northwest, in: Proceedings of the 2006 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an Energy-Efficient Economy, Washington, DC, USA, 2006. [2] R.J. Mowris, A. Blankenship, E. Jones, Field measurements of air conditioners with without TXVs, in: Proceedings of the 2004 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an Energy-Efficient Economy, Washington, DC, USA, 2004. [3] A.A. Domingorena, S.J. Ball, Performance Evaluation of a Selected Three-Ton Air-to-Air Heat Pump in the Heating Mode, ORNL/CON-34, Oak Ridge National Laboratory, Oak Ridge, TN, USA, 1980. [4] D.L. O’Neal, M. Farzad, The effect of improper refrigerant charging on the performance of an air conditioner with capillary tube expansion, Energy and Buildings 14 (1990) 363–371. [5] B. Shen, E.A. Groll, J.E. Braun, Improvement and Validation of Unitary Air Conditioner and Heat Pump Simulation Models at Off-design Conditions: Final Report-1173-RP, American Society of Heating, Refrigerating, and AirConditioning Engineers, Atlanta GA, USA, 2006. [6] AHRI, AHRI Standard 210/240-2006, Standard for Performance Rating of Unitary Air-conditioning and Air-source Heat Pump Equipment, Air-conditioning, Heating, and Refrigeration Institute, Arlington, VA, USA, 2006.
[7] ASHRAE, ASHRAE Standard 116-1995, Methods of Testing for Rating Seasonal Efficiency of Unitary Air-Conditioners and Heat Pumps, American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Atlanta, GA, 1995. [8] V.D. Baxter, J.C. Moyers, Air-Source Heat Pump: Field Measurement of Cycling, Frosting, and Defrosting Losses; 1981-83, ORNL/CON-150, Oak Ridge National Laboratory, Oak Ridge, TN, USA, 1984. [9] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1989 (chapter 28). [10] DOE, 10 CFR Part 430, Energy Conservation Program for Consumer Products: Test Procedure for Residential Central Air Conditioners and Heat Pumps; Final Rule, Department of Energy, USA, 2005. [11] P. Fairey, B. Wilcox, D.S. Parker, M. Lombardi, Climate impacts on heating seasonal performance factor (HSPF) and seasonal energy efficiency ratio (SEER) for air-source heat pumps, ASHRAE Transactions 110 (2004) 178–188. [12] P.W. Francisco, L. Palmiter, D. Baylon, Understanding heating seasonal performance factors for heat pumps, in: Proceedings of the 2004 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an EnergyEfficient Economy, Washington, DC, USA, 2004. [13] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1985 (chapter 28). [14] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1993 (chapter 28).
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Measured effect of airflow and refrigerant charge on the seasonal performance of an air-source heat pump using R-410A Larry Palmiter a , Jun-Hyeung Kim b,∗ , Ben Larson a , Paul W. Francisco c , Eckhard A. Groll d , James E. Braun d a
Ecotope, Inc., Seattle, WA 98105, USA School of Mechanical Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA c Building Research Council, University of Illinois, Urbana-Champaign, IL 61280, USA d School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USA b
a r t i c l e
i n f o
Article history: Received 7 December 2010 Received in revised form 2 February 2011 Accepted 22 March 2011 Keywords: Seasonal performance Heat pumps Refrigerant charge Airflow
a b s t r a c t The objective of this study was to measure the effects of improper airflow and refrigerant charge on the seasonal performance of a typical 10.6 kW, R-410A residential heat pump with a thermostatic expansion valve. Heating and cooling tests were performed in combinations of three refrigerant charges of 75%, 100%, and 125% of nominal value and two airflows of 75% and 100% of rated airflow. In addition, cyclic tests were performed to estimate the heating and cooling seasonal coefficient of performance (COP) at six climate zones specified by Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard 210/240-2008. Results showed that, in each climate zone, increases in refrigerant charge at the rated airflow could improve the unit’s heating seasonal COP by as much as 5%. However, combined decreases in airflow and refrigerant charge could penalize the unit’s heating seasonal COP by as much as 10%. © 2011 Elsevier B.V. All rights reserved.
1. Introduction and background Heat pumps have enjoyed a significant increase in popularity in recent years in the Pacific Northwest, both with the public and with utility program designers. Their popularity is due in part to their ability to save overall electricity throughout the heating season and reduce peak heating loads when compared to electric resistance heating. It is important to note that the savings will only be fully realized for properly installed and maintained equipment. In order to evaluate the benefits of heat pumps in utility programs, it is necessary to understand the degree to which heat pump performance is affected by incorrect charge and airflow and to focus efforts on ensuring that appropriate levels of effort are taken to avoid substantial penalties [1]. Previous studies have demonstrated that incorrect refrigerant charge and improper airflow would cause the performance of air
Abbreviations: Cd , coefficient of degradation; COP, coefficient of performance; IDDB, indoor dry bulb temperature; HSPF, heating seasonal performance factor; ODDB, outdoor dry bulb temperature; ODWB, outdoor wet bulb temperature; PLF, part load factor; SEER, seasonal energy efficiency ratio; TXV, thermostatic expansion valve. ∗ Corresponding author. Tel.: +1 205 348 4492; fax: +1 205 348 6419. E-mail addresses: [email protected] (L. Palmiter), [email protected] (J.-H. Kim), [email protected] (B. Larson), [email protected] (P.W. Francisco), [email protected] (E.A. Groll), [email protected] (J.E. Braun). 0378-7788/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2011.03.026
conditioners to degrade. Mowris et al. [2] reported that about 6 out of 10 air conditioners in the field were operated with improper refrigerant charge and airflow and that this caused the air conditioners to underperform by about 10–20%. Their study strongly suggests that heat pumps in field operation would have a similar problem. Domingorena and Ball [3] measured the effect of refrigerant charge varying from 15% below to 25% above the nominal value on heat pump heating performance. The heating performance remained relatively unchanged at both undercharged and overcharged conditions. They attributed this performance insensitivity to the suction-line accumulator. O’Neal and Farzad [4] studied the effect of improper refrigerant charge on the seasonal cooling performance of a 10.6 kW R-22 air conditioner with capillary tube expansion. They carried out steadystate and cycling cooling performance tests while changing the refrigerant charge from 20% below to 20% above the full charge. The cooling seasonal COP peaked at the full charge of 4.0 kg. However, at 20% undercharged and 20% overcharged conditions, the cooling seasonal COP decreased by 20% and 10%, respectively. Shen et al. [5] measured the cooling performance of a 10.6 kW R410A unitary split heat pump with a thermostatic expansion valve (TXV) under reduced evaporator airflow at dry and wet conditions. Results showed that the cooling performance was degraded much more at dry conditions than at wet conditions. A 20% reduction in evaporator airflow at dry conditions caused 14% and 10% drops in
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
1803
Table 1 Heating mode testing conditions. Test description
Air entering indoor unit temperature Dry-bulb (◦ C)
H1 test (steady) H1C test (cyclic) H3 test (steady) 35 test (steady)a a
21.1 21.1 21.1 21.1
Air entering outdoor unit temperature Wet-bulb (◦ C) max
15.6 15.6max 15.6max 15.6max
Dry-bulb (◦ C) 8.3 8.3 −8.3 1.7
Wet-bulb (◦ C) 6.1 6.1 −9.4 –
Test results at the “35” conditions were used to estimate the defrost penalty at the H2 (frost accumulation) conditions.
cooling capacity and COP, respectively, while the same reduction in evaporator airflow at wet conditions resulted in 5% drops in both cooling capacity and COP. Results also indicated that the cooling performance at wet conditions was relatively unchanged until the evaporator airflow decreased 25%. These studies indicate that improper refrigerant charge and incorrect blower airflow would penalize the performance of heat pumps in both heating and cooling mode; however, their results are mostly limited to steady-state operation. No cycling tests have been performed with heat pumps in heating mode, which is essential for studying the effect of improper refrigerant charge and blower airflow on the heating seasonal COP of heat pumps. Furthermore, no study has been found in the literature examining the combined effect of improper refrigerant charge and blower airflow on the heating seasonal COP of heat pumps. The seasonal performance of heat pumps is measured in terms of heating and cooling seasonal COP values,1 which are set forth by test procedures in the Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard. 210/240 [6]. Since the actual heating performance of heat pumps changes with different climate conditions, AHRI Standard 210/240 [6] specifies six different climate conditions, i.e., Climate Zones 1 through 6. However, only the rated heating seasonal COP of heat pumps, determined based on Climate Zone 4, are normally made available to the public, while the other heating seasonal COP values are not reported. Although the rated heating seasonal COP does not represent the predicted performance in other climate zones other than Climate Zone 4, it is commonly used to estimate electricity use during the heating season in different climate zones. For instance, for Climate Zone 1, which covers most of the area of the state of Florida in the U.S., the use of a rated heating seasonal COP would overestimate the actual electricity use during the heating season. Thus, in order to allow for climate-related variations in heat pump performance, it is necessary to determine the heating seasonal COP values of heat pumps in all six climate zones and to use the proper heating seasonal COP value for each climate zone in estimating the annual heating costs. The goal of this study is to determine the impact of refrigerant charge and blower airflow variations on the seasonal performance of a 10.6 kW R-410A residential heat pump. This study does not specifically attempt to determine the rated performance of the unit, but rather to measure the expected in situ performance which could result from a variety of typical installation practices [1]. All tests were performed by closely following the methods used in the rating standards to establish a common base for performance comparisons. All test procedures and results for the heat pump are summarized in this paper, and the effects of improper refrigerant charge and blower airflow on the seasonal heating and cooling performance are then discussed for all six climate zones in the U.S.
1 Heating and cooling seasonal COP values are determined by dividing the heating seasonal performance factor (HSPF) and seasonal energy efficiency ratio (SEER) each by 3.412 Btu/Wh.
2. Test method The medium-performance heat pump tested was a 10.6 kW R-410A model with a scroll compressor and a suction-line accumulator. The outdoor unit had a one-row, 787-fins-per-meter coil with a face area of 1.87 m2 . The compressor was rated at 79 locked rotor amps and 17.1 rated load amps. The outdoor fan was a propellertype, direct-drive fan rated at 0.9 full load amps. The indoor unit had a three-row, 570.9-fins-per-meter coil with a face area of 0.32 m2 . The indoor fan was a centrifugal-type, direct-drive fan. The heat pump included both a time-delay relay and a TXV with a charging subcooling of 5.6 K. As the project was to measure and compare the system performance of a new, factory-provided heat pump at different charge levels and airflow rates, the unit was purchased from a local distributor and installed according to the installation manual with minimum intervention. The only exception to a completely standard setup process was the charge adjustment to properly account for the installed liquid refrigerant line length per the installation instructions. All laboratory tests were done to characterize the performance at the rating conditions specified in Air-Conditioning, Heating, and Refrigeration Institute (AHRI) Standard 210/240 [6]. A set of tests at extremely hot conditions such as those found in the Southwest U.S. desert was also done. The tests were conducted at each of three outdoor temperatures in heating mode (−8.3 ◦ C, 1.7 ◦ C, and 8.3 ◦ C) and in cooling mode (27.8 ◦ C, 35 ◦ C, and 51.7 ◦ C). These tests provide useful comparison to the manufacturer’s published heating and cooling seasonal COP values for the unit. Additionally, in order to measure off-design performance in both modes, refrigerant charges were varied from approximately 25% above to 25% below the manufacturer’s nominal recommended charge at each 75% and 100% of rated airflow. The airflow was varied using a booster fan in the laboratory, and the airflow adjustments in this study were specifically designed to target the effects resulting from a wide range of different conditions leading to different static pressures. In practice, many heat pump applications have flow rates lower than the nominal airflow rate, which can lead to problems with both cooling and heating performance. In heating mode, capacity, power, fan power and COP were measured in all combinations of the three outdoor temperatures of −8.3 ◦ C, 1.7 ◦ C, and 8.3 ◦ C, three refrigerant charges of 75%, 100%, and 125% of the nominal value, and two airflows of 0.4248 and 0.566 m3 /s (corresponding to 75% and 100% of rated airflow). The detailed heating mode test conditions are listed in Table 1. Partload performance tests were also conducted to characterize the coefficient of degradation (Cd ), which is essential for calculating the heating seasonal COP. The Cd measurement was made only at a temperature of 8.3 ◦ C (H1 steady and H1C cyclic test conditions). Cooling mode tests were performed at the cooling mode test conditions listed in Table 2. Understanding the cooling mode tests requires a more detailed discussion of refrigerant charge levels. The catalog-recommended charge is 3.36 kg of R-410A. The test setup had an extra 3.05 m of refrigerant line. Per manufacturer’s
1804
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Table 2 Cooling mode testing conditions. Test description
A test (steady, wet coil) B test (steady, wet coil) C test (steady, dry coil) D test (cyclic, dry coil) 125 testa (steady, dry coil) 120 testb (steady, dry coil) a b
Air entering indoor unit temperature
Air entering outdoor unit temperature
Dry-bulb (◦ C)
Wet-bulb (◦ C)
Dry-bulb (◦ C)
Wet-bulb (◦ C)
26.7 26.7 26.7 26.7 26.7 26.7
19.4 19.4 – – – –
35 27.8 27.8 27.8 51.7 48.9
– – – –
The 125 test conditions is added for hot desert test conditions in the Southwest U.S. For 117% charge, the 120 test conditions were used in replacement of the 125 test conditions due to high-side pressure switching.
Table 3 Performance comparison of laboratory tests with catalog data at 3.54 kg charge level and 0.5663 m3 /s. Test
B A 125 a b
Testing results/catalog dataa
Testing/catalog ratio data
Capacity (kW)
Power (kW)
COP
Cooling seasonal COPb
Capacity (%)
Power (%)
COP (%)
Cooling seasonal COP (%)
10.43/10.62 9.89/9.91 7.14/7.99
2.95/2.78 3.37/3.17 4.95/4.29
3.54/3.82 2.93/3.12 1.44/1.86
3.47/3.81
98.2 99.8 89.3
106.1 106.3 115.4
92.6 93.9 77.4
91.0
Capacity and power interpolated from table for the B test. Cooling seasonal COP calculated using a coefficient of degradation, Cd = 0.04, and a part load factor, PLF = 0.98.
recommendations, the test laboratory added 0.19 kg of refrigerant to compensate for the extra length resulting in a base charge (100%) of 3.54 kg. The high charge (125%) was then 4.43 kg and the low charge (75%) was 2.66 kg. All of the heating mode tests were done with these values, and cooling mode tests were done at 100% and 75% charge. It was planned to perform an additional set of tests at an outdoor temperature of 51.7 ◦ C (125 test condition) with dry indoor coil to examine the performance under hot, dry conditions. Unfortunately, at the 125% charge level, the high-pressure limit switched the compressor off at outdoor temperatures above 29.4 ◦ C. As a result, it was decided to run the high-charge tests at only 3.94 kg of charge, about 111% of the 3.54 kg base level. In the first base charge tests, it was noticed that the test results did not compare well with the catalog data. Table 3 shows the comparison with catalog data for the first base charge tests. In comparison to the catalog data, the test results show lower capacity and higher power resulting in a lower COP, especially at 51.7 ◦ C (125 test conditions), where the COP was 23% lower. The seasonal cooling COP was 9% below the catalog rating. In addition, the measured subcooling was 8.4 K, which is 2.8 K higher than the rated subcooling of 5.6 K. It was suggested that the addition of the extra 0.19 kg of refrigerant might have caused the relatively large difference between the experimental data and the catalog data. All of the base charge tests were redone at a charge of 3.36 kg, where the measured and rated subcoolings were matched. Table 4 compares these tests with the catalog data. The agreement is much better, especially at 51.7 ◦ C. The power comparisons are now very close. The cooling seasonal COP is now about 6.5% below the catalog rating. This unit is very sensitive to additional charge, even in a situation where the length of line set would jus-
tify the extra charge. The recommended charge is close to the upper limit that allows heat pump cooling operation at 51.7 ◦ C outdoor temperature. This may indicate the need for precautions to avoid overcharging this unit in hot climates. Due to the matched subcooling and better performance agreement, it was decided to use 3.36 kg as the base level (100%) for the cooling tests. The charge level of 2.66 kg was then at 79% of the base and the charge level of 3.94 kg was 117% of the base. Also, the hightemperature tests were done at an outdoor temperature of 48.9 ◦ C instead of 51.7 ◦ C for the high charge tests. As a result, cooling mode tests were performed for all six combinations of refrigerant charges of 79%, 100%, and 117% of the nominal value (3.36 kg of R-410A), and air flows of 0.4248 and 0.5663 m3 /s (corresponding to 75% and 100% of rated airflow). For each of the six combinations, steady-state tests were done at three different outdoor temperatures (27.8 ◦ C, 35 ◦ C, and 51.7 ◦ C, except for the high charge case where the outdoor temperature was 48.9 ◦ C). In addition, cyclic tests were carried out at an outdoor temperature of 27.8 ◦ C with a dry indoor coil to determine the Cd and subsequently the cooling seasonal COP. The heat pump tests were designed in accordance with AHRI Standard 210/240 [6]. For each combination of airflow, refrigerant charge, and outdoor temperature, a single test of the equipment was conducted. With the steady-state tests, the environmental chambers and the unit were operated until they reached the steady state conditions, and then the test data were recorded. For the cyclic tests, the unit was cycled on and off at least two times, and then immediately following the cycling, the test data were recorded. A number of measurements were recorded at each testing point including airflow, dew point, unit power consumption, air and
Table 4 Performance comparison of laboratory tests with catalog data at 3.36 kg charge level and 0.5663 m3 /s. Test
B A 125 a b
Testing results/catalog dataa
Testing/catalog ratio data
Capacity (kW)
Power (kW)
COP
Cooling seasonal COPb
Capacity (%)
Power (%)
COP (%)
Cooling seasonal COP (%)
10.50/10.62 9.67/9.91 7.34/7.99
2.83/2.78 3.20/3.17 4.42/4.29
3.71/3.82 3.02/3.12 1.66/1.86
3.56/3.81
98.8 97.6 91.8
101.8 100.9 103.0
97.1 96.7 89.1
93.5
Capacity and power interpolated from table for the B test. Cooling seasonal COP calculated using a coefficient of degradation, Cd = 0.08, and a part load factor, PLF = 0.96.
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
1805
Table 5 Normalization values of heating capacity and COP at nominal charge and airflow. Test condition
Capacity (kW)
COP
H1 (Tout = 8.3 ◦ C) 35 (Tout = 1.7 ◦ C) H3 (Tout = −8.3 ◦ C)
10.71 8.87 6.24
3.51 3.08 2.31
3.1. Heating capacity
Fig. 1. Heating capacity ratio versus airflow rate at the H3, 35, and H1 test conditions.
refrigerant temperatures and pressures, and refrigerant mass flow rate. • Airflow was measured with a nozzle apparatus, built in accordance with ASHRAE standard 116 [7] and having an uncertainty of ±10%. • The hygrometer to measure dew point was accurate within ±0.2 ◦ C. • For temperature measurements in both the air and the refrigerant, T-type thermocouples with an error of ±1 ◦ C were employed. For the air-side measurements, eight thermocouples were arranged in a grid placed in the flow path and averaged to determine the temperature. • Seven pressure transducers were installed to measure pressure on the refrigerant side. Three had an error of ±0.13% of full scale and four had an error of ±1.0% of full scale. • A Coriolis flow meter, with an error of ±0.5% of full scale, was installed in the liquid line to measure refrigerant mass flow rate. • To measure indoor blower and outdoor fan power consumptions and compressor power consumption, digital watt transducers with an accuracy of ±0.2% were used. 3. Heating mode test results This section summarizes the effects of refrigerant charge and airflow on capacity, COP, Cd , and the defrost penalty. Figs. 1 and 2 show the capacity and COP ratios, respectively, at the H3, 35, and H1 test conditions. Under each test conditions, capacities and COP values are normalized to those measured at 100% refrigerant charge (3.54 kg) and 0.5663 m3 /s airflow (100% of rated airflow). Table 5 shows the capacities and COP values used to normalize each point.
Fig. 1 shows the effect of airflow on capacity for each of the nine combined cases of test conditions and charge level (e.g., in the case of H3-75% charge, H3 indicates test conditions, and 75% charge shows charge level). In all nine cases, the capacity decreases about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). Between the nine cases at 0.5663 m3 /s (100% of rated airflow), the 75% charge level reduces capacity by approximately 3% and 8% at the 35 (1.7 ◦ C) and H1 (8.3 ◦ C) test conditions, respectively, compared to their 100% charge levels. However, at the H3 (−8.3 ◦ C) test conditions, the 75% charge level actually increases capacity by 5%. On the other hand, the 125% charge level does not affect capacity at either the 35 (1.7 ◦ C) or H1 (8.3 ◦ C) test conditions, but improves capacity by 3% at the H3 (−8.3◦ C) test conditions. 3.2. Coefficient of performance for heating The coefficient of performance (COP) is also calculated using AHRI Standard 210/240 [6], which includes the indoor blower and outdoor fan power as well as that of the compressor in the denominator. The heat generated by the indoor fan is included in the capacity in the numerator. Each of these values was measured in the laboratory. AHRI mandates that for test units that do not include a specific indoor air handler, a default value for the indoor blower power of 365 W per 0.4719 m3 /s must be used. The indoor fan power measurements for the model tested averaged about 391 W at 0.4248 m3 /s (75% of rated airflow) and 533 W at 0.5663 m3 /s (100% of rated airflow). The outdoor fan measurements for the current tests averaged 212 W. Fig. 2 shows the effect of airflow on heating COP for each of the nine combined cases of test conditions and charge level (e.g., in the case of H3-75% charge, H3 indicates the test conditions, and 75% charge shows the charge level). In all nine cases, the COP decreases by about 6% for a 25% reduction in airflow. The effect of charge on COP is within about ±3%. Figs. 1 and 2 clearly show that the unit should not be run with airflow below the nominal airflow due to the performance penalties. 3.3. Part-load and defrost penalty factors
Fig. 2. Heating COP ratio versus airflow rate at the H3, 35, and H1 test conditions.
Two other factors have a large effect on heat pump performance: part-load operation and the defrost cycle. Under part-load conditions, the heat pump cycles off and on. For a short period during the start-up the heat pump draws nearly full power, but there is no output while the appropriate equilibrium conditions are being established throughout the refrigerant side. The indoor fan is off during this period. Additionally, each time the unit cycles off there is heat loss in the system. The net effect of these losses is an increasing loss of efficiency as the unit runs for a smaller fraction of time. The coefficient of degradation (Cd ) is the efficiency loss that occurs due to the cycling of the unit. AHRI Standard 210/240 [6] assumes a linear percentage loss between zero load and full load (where the efficiency reaches the steady-state value). For instance, with a Cd of 0.25 the efficiency at zero load is reduced to 75% of the steady-
1806
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Fig. 3. Cd versus airflow rate.
state value, and at 50% load the efficiency is reduced to 87.5% of the steady-state value. An additional performance loss occurs under outdoor conditions that lead to ice buildup on the outdoor coil. During the defrost cycle the outdoor fan is off, and the heat pump operates as an air-conditioner to warm the outdoor coils. In order to avoid occupant discomfort, typically the indoor fan and backup heat also run during this cycle to warm the air coming off the cold indoor coil. Unfortunately, due to contract deadlines, the defrost tests at the H2 conditions (1.7 ◦ C ODDB, 0.6 ◦ C ODWB, and 21.1 ◦ C IDDB) were not conducted. However, the defrost penalty can be stated as a multiplier of the steady-state efficiency measured at the 35 test conditions [8]. A multiplier of 0.9 [9] was used to estimate the penalized heating capacity and COP at the H2 conditions for the test unit, and the estimated heating capacity and COP were used for the heating seasonal COP calculations. The AHRI Standard requires the measurement of Cd and the defrost penalty, but neither heat pump manufacturers nor AHRI publish these measured values. Although unpublished, they are used by the manufacturer to calculate the heating seasonal COP. In this study, one of the goals is to perform all of the required measurements needed to calculate the heating seasonal COP for various combinations of airflow and charge level. Fig. 3 illustrates the effect of airflow and charge on the Cd . It shows that the impact of airflow on the Cd is only a few percent at each charge level, and the Cd at 75% charge is about 0.3, at nominal charge about 0.25 and at 125% charge about 0.16. However, the impact of charge on the Cd is significant. The Cd at 75% charge is about 20% higher than that at nominal charge while the Cd at 125% charge is about 36% lower. This suggests that the heat pump can produce heating at each start-up more quickly when it is over-charged to some extent. 4. Heating seasonal performance It is difficult to estimate the seasonal performance of a heat pump because both the capacity and the COP depend strongly on outdoor temperature. In addition, the capacity at low outdoor temperatures is usually not adequate to meet the heating load, thus requiring the use of backup heat. Also, the effects of part-load operation and defrost must be taken into account. In the late 1970s a bin-hour calculation method that accounts, to some extent, for all of these effects was developed. This method results in the heating seasonal COP.2 According to AHRI Standard 210/240 [6], Section 4.2, heat pump manufacturers are required to calculate the heating seasonal COP for given bin-hour profiles for six different climate
2 Heating seasonal COP is determined by dividing the heating seasonal performance factor (HSPF) by 3.412 Btu/Wh.
zones “unless an approved alternative rating method is used, as set forth in 10 CFR 430.24(m), Subpart B [10].” However, the U.S. government only allows the heating seasonal COP for Climate Zone 4 to appear on the label. This is often misleading for performance in other regions because the heating seasonal COP varies strongly with climate zone and heating load [11]. Climate and heating load differences also make it so that the relative benefits of a more efficient heat pump compared to a less efficient heat pump are not consistent between climate zones [12]. For example, a heat pump with a 10% higher published (Climate Zone 4) heating seasonal COP than another will not result in a 10% efficiency improvement in all climates, with some climates having lesser benefits and some climates having greater benefits. The laboratory test data were used in this study for calculating the heating seasonal COP in accordance with AHRI Standard 210/240 [6] for all six climate zones. 4.1. Calculation method AHRI Standard 210/240 [6] defines the test methods used to measure the heating seasonal COP of a heat pump as well as the equations required for calculation. The necessary variables for the calculation are airflow rates, heating capacities and electrical power consumptions from steady-state tests at −8.3 ◦ C (H3 test conditions), 1.7 ◦ C (35 test conditions), and 8.3 ◦ C (H1 test conditions), a cyclic test at 8.3 ◦ C, and a defrost test at 1.7 ◦ C. The capacity and input power at other outdoor temperatures are estimated as follows. In the capacity and input power versus outdoor temperature plots, the slope of the line below −8.3 ◦ C and above 7.2 ◦ C is equal to that of a line connecting the 1.7 ◦ C and 8.3 ◦ C test point values. The central portions of the curves are defined by connecting the −8.3 ◦ C and 1.7 ◦ C test values and the newly defined point at 7.2 ◦ C. The load, capacity, compressor input, and auxiliary heat (assumed all-electric) are then calculated for each temperature bin, applying part-load corrections as needed. There are some assumptions implicit in this method. It assumes that there is no defrost penalty below −8.3 ◦ C; however, for units with time–temperature defrost control there will be significant defrost penalties at all temperatures below about 4.4 ◦ C. There is also a large discontinuity in the performance curves at 7.2 ◦ C, an effect that does not occur in the laboratory. Additionally, the house load assumes a home heating balance point of 18.3 ◦ C (meaning the mimimum outdoor temperature at which no heat is required by the home), which is on the high side for well-insulated homes. This is compensated somewhat by multiplication of the load at each outdoor temperature by a load factor of 0.77. The load factor procedure, however, is no longer sanctioned by ASHRAE. ASHRAE recognized the problem of using the load factor as early as the 1985 Handbook of Fundamentals [13] and by the 1993 Handbook of Fundamentals [14] the load factor was no longer included in the discussion of energy estimating methods. Instead the variable-base degree day method was used. Nevertheless, heating seasonal COP serves as a single point rating for heat pump performance that attempts to account for all of the major factors affecting performance. The largest problem lies in failing to publish the ratings for all six zones (although at least one major manufacturer publishes heating seasonal COP ratings for Climate Zone 5 in addition to Climate Zone 4). 4.2. Heating seasonal COP bin data Fig. 4 shows the fractional temperature bin data from AHRI Standard 210/240 [6] used for the six climate zones. The zones pertinent to the Pacific Northwest are 4, 5, and 6. The climates of Boise, ID and Spokane, WA are well represented by Climate Zone 4 (the rating Climate Zone), while Missoula, MT is in Climate Zone 5. Climate Zone 6 is representative of the climate west of the Cascade Mountains in
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Fig. 4. Fractional bin hour data for heating seasonal COP calculations (AHRI Standard 210/240 [6]). Table 6 Heating seasonal COP at nominal charge and airflow. Zone
1
2
3
4
5
6
Heating seasonal COP
2.69
2.63
2.50
2.29
1.95
2.64
Oregon and Washington, e.g., Seattle, WA and Portland, OR. It has a peak in the relative bin temperature distribution at about 10 ◦ C and very few hours at cold temperatures. It should be noted that the Climate Zone map of the US provided in the AHRI Standard 210/240 [6] was incorrectly drawn in the first publication of the Standard and, unfortunately, has never been corrected. This error results in the assignment of completely incorrect climate zones for the region west of the Cascade Mountains in Oregon, Washington, and Northern California. For example, Seattle lies on the border between Climate Zones 4 and 5 and Portland is fully in Climate Zone 5 in the map. This area west of the Cascades is actually Climate Zone 6, which has a level of heat pump performance in heating mode comparable with that in Climate Zone 1.
1807
Fig. 6. Normalized heating seasonal COP versus airflow rate in zones 4, 5, and 6.
Figs. 5 and 6 show the effect of airflow and charge on heating seasonal COP for each of the six climate zones. The heating seasonal COP decreases by about 3–4% when the airflow is reduced from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for each climate zone and charge level. Reducing the charge to 75% results in an 8% loss in heating seasonal COP and increasing the charge to 125% results in an 8% gain for the warmer climate zones (1, 2, 3 and 6). For the colder climates, the effect of a 25% change in charge is about 3–5%. 5. Cooling mode results
Table 6 shows heating seasonal COP values calculated from the laboratory test data for nominal charge and airflow. These values were used to normalize the heating seasonal COP plots in Figs. 5 and 6. It should be noted that there is a large variation in heating seasonal COP across climate zones ranging from the lowest value of 1.95 in Zone 5 to the largest value of 2.69 in Zone 1. The measured value of 2.29 in Zone 4 shows excellent agreement with the published heating seasonal COP for this heat pump of 2.32.
In cooling mode, the standard tests required by AHRI Standard 210/240 [6] were performed. These tests were made at outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions) with wet indoor coil (entering wet bulb temperature of 19.4 ◦ C and dry bulb of 26.7 ◦ C). In order to normalize the performance graphs, the measured capacity and COP at 0.5663 m3 /s (100% of rated airflow) and at 3.36 kg of charge (100% charge) in Table 4 were used. As expected, the capacity and COP decrease rapidly with increasing outdoor temperature. Fig. 7 shows the normalized cooling capacity ratios as a function of airflow and charge level at each outdoor temperature. The capacity decreases by about 6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions). For the highest outdoor temperatures, the airflow effect was about 10%. At 79% charge, the capacity is reduced 10% or more for the lower two temperatures across all airflow levels. For the highest outdoor temperatures at 79% charge the capacity is reduced over 10% for 0.4248 m3 /s airflow (75% of rated airflow).
Fig. 5. Normalized heating seasonal COP versus airflow rate in zones 1, 2, and 3.
Fig. 7. Cooling capacity ratio versus airflow rate at the B, A, and 125 (or 120) test conditions.
4.3. Effect of charge and airflow on heating seasonal COP
1808
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
6. Conclusions
Fig. 8. Cooling COP ratio versus airflow rate at the B, A, and 125 (or 120) test conditions.
Results show that the heating seasonal COP of the heat pump differs widely between the six climate zones in the U.S, ranging from 1.95 in Zone 5 to 2.69 in Zone 1, while it is 2.29 in Zone 4. Results also show that, in each climate zone, different airflow and refrigerant charge for the heat pump would make a noticeable impact on the unit’s seasonal performance (i.e., heating and cooling COP). For instance, increased charge tends to have substantial benefits in heating mode. However, the sharp performance drop in cooling mode, including shutting down due to the triggering of the high pressure limit switch at high temperatures, indicates that caution must be taken with regard to overcharging a unit if cooling is of substantial interest. Decreased airflow tends to produce significant seasonal performance penalties in both heating and cooling mode. This shows the importance of proper airflow during the heating and cooling seasons. 6.1. Heating mode The major findings of the heating mode laboratory tests and the heating seasonal COP calculations are summarized below.
Fig. 9. Cooling seasonal COP ratio versus airflow rate.
Fig. 8 shows the normalized cooling COP ratios at two different airflows. At 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions), the COP decreases by about 2% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). At high temperatures of 48.9 ◦ C (120 test conditions) and 51.7 ◦ C (125 test conditions), the COP is penalized by 4–10% as airflow decreases. Fig. 8 also shows that overcharging or undercharging the unit results in COP decreases of at least 8% for these temperatures at each airflow. Figs. 9 and 10 show the impact of charge and airflow on the cooling seasonal COP and the coefficient of degradation Cd , respectively. At each refrigerant charge, changes in airflow do not make a noticeable impact on the cooling seasonal COP. However, increasing or reducing charge changes the cooling seasonal COP with reductions of 3–4% for high-charge and 10% for low-charge at each airflow rate. The Cd demonstrates a strong dependence on charge. At nominal airflow it ranges from near 0.00 at 117% charge to 0.09 at 79%. The effect of airflow can be seen at 79 and 100% charge with Cd ranging from 0.07 to 0.09. Airflow has no significant effect at 117% charge.
Fig. 10. Coefficient of degradation versus airflow rate.
• Heat pump capacity decreases about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). The most significant impact of reduced charge (8–10%) is at 8.3 ◦ C where the loss in capacity will have little impact due to small heating loads. • Heat pump COP shows decreases of about 5–6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) at each charge level and temperature. In contrast, the impact of charge on COP is quite small (1–3%). • The coefficient of degradation, Cd , at 75% charge is about 0.3 and at 100% charge about 0.25. The impact of airflow is only a few percent. • The calculated heating seasonal COP at nominal conditions varies noticeably across climate zones from 1.95 in Zone 5 to 2.69 in Zone 1. The value of 2.29 calculated for Zone 4 shows excellent agreement with the published value for this heat pump of 2.32. • The heating seasonal COP decreases by about 3 to 4% when the air flow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for each climate zone and charge level. Reducing the charge to 75% results in an 8% loss in heating seasonal COP, and increasing the charge to 125% results in an 8% gain for the warmer climate zones (1, 2, 3 and 6). For the colder climates, the effect of a 25% change in charge is about 3–5%. 6.2. Cooling mode The major findings for cooling mode are summarized below. In cooling mode the performance is more strongly affected by both airflow and charge. • Difficulties in operating the equipment were encountered at high temperatures with even moderate levels of overcharge. This stresses the need for precautions to prevent overcharging, especially when the unit is used in hot climates. • The capacity decreases by about 6% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow) for outdoor temperatures of 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions). For the highest outdoor temperatures, the airflow effect is about 10%. • At 79% charge, the capacity is reduced 10% or more for the lower two temperatures across all airflow levels. For the highest out-
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
•
•
•
•
door temperatures at 79% charge, the capacity is reduced over 10% for 0.4248 m3 /s airflow (75% of rated airflow). At 27.8 ◦ C (B test conditions) and 35 ◦ C (A test conditions), COP decreases by about 2% when the airflow decreases from 0.5663 m3 /s (100% of rated airflow) to 0.4248 m3 /s (75% of rated airflow). At high temperatures of 48.9 ◦ C (120 test conditions) and 51.7 ◦ C (125 test conditions), the COP is penalized by 4–10% as the airflow decreases. Overcharging or undercharging the unit results in a COP decrease of at least 8% for these temperatures. The effect of overcharging the unit significantly reduces the COP. At high charge, the COP decreases by at least 10% compared with that at nominal charge for all airflows. The cooling seasonal COP value shows a large dependence on charge level but little on airflow. 117% charge reduces the cooling seasonal COP by 3–4% while 79% charge reduces by 10%. The Cd also demonstrates a strong dependence on charge. At nominal airflow, it ranges from near 0 at 117% charge to 0.09 at 79% and 100% charge, which is less than the default Cd value of 0.25. Airflow has little effect on the Cd at all charge levels. The coefficient of degradation, Cd , at 79% and 100% charge ranges from 0.07 to 0.09. The Cd at 117% charge is zero.
1809
pump performance in general. In particular, a heat pump without a suction-line accumulator may be much more sensitive to variations in charge level. However, investigations of the impact of the accumulator relative to its absence were not within the scope of this study. These results only apply to the heat pump tested and should not be generalized across other heat pump models, even those using the same refrigerant. Additional testing is needed to identify the range of results possible across all available heat pumps. However, the results are examples of conditions that can be found in at least one heat pump model.
Acknowledgements This research was funded by the US Department of Energy, the Bonneville Power Administration, and the States of Oregon, Washington, Idaho, and Montana. The project manager was Ken Eklund of the Idaho Energy Division.
Appendix A. 7. Limitations and suggestions The results suggest that caution should be taken in making any statements about the effects of charge and airflow on heat
Performance test data for the 10.6 kW heat pump including cooling and heating capacities, compressor powers, fan powers, and COPs are tabulated (Tables A.1–A.4).
Table A.1 Cooling mode test data at steady-state conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
A A B B C C 125 125
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
2.66 2.66 2.66 2.66 2.66 2.66 2.66 2.66
A A B B C C 125 125
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
A A B B C C 120 120
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
Indoor net cooling (kW)
Indoor sensible cooling (kW)
COP (–)
Indoor fan power (W)
Outdoor fan power (W)
Outdoor compressor power (W)
8.29 8.79 8.81 9.36 7.97 8.78 6.26 6.71
6.12 7.13 6.39 7.35 7.97 8.78 6.26 6.71
2.77 2.81 3.33 3.37 3.00 3.16 1.52 1.58
396 523 397 523 403 526 400 529
196 195 198 198 201 197 195 195
2405 2405 2051 2058 2049 2051 3529 3534
3.36 3.36 3.36 3.36 3.36 3.36 3.36 3.36
9.08 9.67 9.78 10.50 8.57 9.52 6.53 7.34
6.53 7.56 6.83 7.95 8.57 9.52 6.53 7.34
2.97 3.02 3.62 3.71 3.21 3.40 1.51 1.66
384 513 387 510 386 510 382 509
191 193 195 195 197 196 187 187
2484 2490 2118 2123 2089 2095 3744 3726
3.94 3.94 3.94 3.94 3.94 3.94 3.94 3.94
9.17 9.77 9.83 10.52 8.42 9.53 6.59 7.39
6.62 7.66 6.87 7.96 8.42 9.53 6.59 7.39
2.70 2.78 3.34 3.44 2.87 3.10 1.33 1.47
384 510 386 510 390 519 384 510
192 192 195 195 196 196 190 189
2818 2815 2361 2356 2350 2357 4388 4324
Table A.2 Cooling mode test data at cyclic conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
Cyclic indoor ave. net cooling (kW)
Cyclic indoor ave. fan power (W)
Cyclic outdoor ave. fan power (W)
Cyclic outdoor ave. comp power (W)
Cyclic average COP (–)
Cyclic coefficient degradation (–)
D D
0.4248 0.5660
2.66 2.66
1.57 1.72
104 136
40 39
410.2 411.4
2.83 2.93
0.07 0.09
D D
0.4248 0.5660
3.36 3.36
1.66 1.84
101 128
39 39
408.0 413.7
3.03 3.17
0.07 0.08
D D
0.4248 0.5660
3.94 3.94
1.64 1.84
102 133
39 39
411.0 415.4
2.98 3.14
0 0
1810
L. Palmiter et al. / Energy and Buildings 43 (2011) 1802–1810
Table A.3 Heating mode test data at steady-state conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
2.66 2.66 2.66 2.66 2.66 2.66
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
H1 H1 H3 H3 35 35
0.4248 0.5663 0.4248 0.5663 0.4248 0.5663
Indoor coil net heating (kW)
COP (–)
Indoor fan power (W)
Outdoor fan power (W)
Outdoor compressor power (W)
9.50 9.91 6.08 6.58 8.21 8.63
3.19 3.38 2.25 2.40 2.86 3.02
387 529 394 538 392 532
209 210 216 218 211 211
2385 2189 2088 1991 2272 2112
3.54 3.54 3.54 3.54 3.54 3.54
10.06 10.71 6.00 6.24 8.23 8.87
3.23 3.51 2.25 2.31 2.86 3.08
389 533 393 531 386 531
209 210 215 216 210 211
2519 2311 2063 1951 2285 2137
4.43 4.43 4.43 4.43 4.43 4.43
9.98 10.76 6.10 6.45 8.42 8.83
3.21 3.48 2.27 2.37 2.89 3.05
389 528 398 538 391 538
210 211 217 219 211 214
2506 2349 2067 1964 2307 2146
Table A.4 Heating mode test data at cyclic conditions. Test conditions
Air flow (m3 /s)
System charge (kg)
Cyclic indoor ave. net heating (kW)
Cyclic indoor ave. fan power (W)
Cyclic outdoor ave. fan power (W)
Cyclic outdoor ave. comp power (W)
Cyclic average COP (–)
Cyclic coefficient degradation (–)
H1C H1C
0.4248 0.5663
2.66 2.66
1.38 1.45
104 138
42 42
427.6 403.8
2.40 2.49
0.29 0.31
H1C H1C
0.4248 0.5663
3.54 3.54
1.55 1.65
105 142
42 42
450.3 419.2
2.60 2.73
0.23 0.26
H1C H1C
0.4248 0.5663
4.43 4.43
1.74 1.88
102 137
42 42
471.2 450.5
2.83 2.98
0.14 0.18
References [1] B. Davis, D. Baylon, S. Strand, It sounded good on paper: field performance of residential heat pumps in the Pacific Northwest, in: Proceedings of the 2006 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an Energy-Efficient Economy, Washington, DC, USA, 2006. [2] R.J. Mowris, A. Blankenship, E. Jones, Field measurements of air conditioners with without TXVs, in: Proceedings of the 2004 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an Energy-Efficient Economy, Washington, DC, USA, 2004. [3] A.A. Domingorena, S.J. Ball, Performance Evaluation of a Selected Three-Ton Air-to-Air Heat Pump in the Heating Mode, ORNL/CON-34, Oak Ridge National Laboratory, Oak Ridge, TN, USA, 1980. [4] D.L. O’Neal, M. Farzad, The effect of improper refrigerant charging on the performance of an air conditioner with capillary tube expansion, Energy and Buildings 14 (1990) 363–371. [5] B. Shen, E.A. Groll, J.E. Braun, Improvement and Validation of Unitary Air Conditioner and Heat Pump Simulation Models at Off-design Conditions: Final Report-1173-RP, American Society of Heating, Refrigerating, and AirConditioning Engineers, Atlanta GA, USA, 2006. [6] AHRI, AHRI Standard 210/240-2006, Standard for Performance Rating of Unitary Air-conditioning and Air-source Heat Pump Equipment, Air-conditioning, Heating, and Refrigeration Institute, Arlington, VA, USA, 2006.
[7] ASHRAE, ASHRAE Standard 116-1995, Methods of Testing for Rating Seasonal Efficiency of Unitary Air-Conditioners and Heat Pumps, American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Atlanta, GA, 1995. [8] V.D. Baxter, J.C. Moyers, Air-Source Heat Pump: Field Measurement of Cycling, Frosting, and Defrosting Losses; 1981-83, ORNL/CON-150, Oak Ridge National Laboratory, Oak Ridge, TN, USA, 1984. [9] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1989 (chapter 28). [10] DOE, 10 CFR Part 430, Energy Conservation Program for Consumer Products: Test Procedure for Residential Central Air Conditioners and Heat Pumps; Final Rule, Department of Energy, USA, 2005. [11] P. Fairey, B. Wilcox, D.S. Parker, M. Lombardi, Climate impacts on heating seasonal performance factor (HSPF) and seasonal energy efficiency ratio (SEER) for air-source heat pumps, ASHRAE Transactions 110 (2004) 178–188. [12] P.W. Francisco, L. Palmiter, D. Baylon, Understanding heating seasonal performance factors for heat pumps, in: Proceedings of the 2004 ACEEE Summer Study on Energy Efficiency in Buildings, American Council for an EnergyEfficient Economy, Washington, DC, USA, 2004. [13] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1985 (chapter 28). [14] American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Handbook of Fundamentals, ASHRAE, Atlanta, GA, 1993 (chapter 28).