EFFECT OF REFRIGERANT CHARGE, DUCT LEAKAGE, AND EVAPORATOR Am FLOW ON THE HIGH TEMPERATURE PERFORMANCE OF AIR CONDITIONERS AND HEAT PUMPS A Thesis by ANGEL GERARDO RODRIGUEZ Submitted to the Office of Graduate Studies of Texas A&M University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE August 1995 Major Subject: Mechanical Engineering
EFFECT OF REFRIGERANT CHARGE, DUCT LEAKAGE, AND EVAPORATOR AIR FLOW ON THE HIGH TEMPERATURE PERFORMANCE OF AIR CONDITIONERS AND HEAT PUMPS A Thesis by ANGEL GERARDO RODRIGUEZ Submitted to the Office of Graduate Studies of Texas A&M University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE
Ill ABSTRACT Effect of Refrigerant Charge, Duct Leakage, and Evaporator Air Flow on the High Temperature Performance of Air Conditioners and Heat Pumps. (August 1995) Angel Gerardo Rodriguez, B.M.E., Georgia Institute of Technology Chair of Advisory Committee: Dr. D.L. O'Neal An experimental study was conducted to quantify the effect of several installation items on the high outdoor ambient temperature performance of air conditioners. These installation items were: improper amount of refrigerant charge, reduced evaporator airflow, and return air leakage from hot attic spaces. There were five sets of tests used for this research: two of them for the charging tests, two for the reduced evaporator airflow, and one for the return air leakage tests. For the charging tests, the indoor room conditions were 80 F (27.8 C) dry-bulb and 50% relative humidity. The outdoor conditions ranged from 95 F (35 C) all the way up to 120 F (48.9 C). Charge levels ranged from 30% undercharged to 40% overcharged for the short-tube orifice unit. For the thermal expansion valve (TXV) unit, charge levels ranged from -36% charging to +27% charging. Performance was quantified with the following variables: total capacity, energy efficiency ratio (EER), and power. The performance of the orifice unit was more sensitive to charge than it was for the TXV unit. For the TXV unit on the -27% to +27% charging range, the capacity and EER changed little with charge. A TXV unit and a short-tube orifice unit were also tested for reduced evaporator airflow. As evaporator airflow decreased, the capacity and EER both decreased as expected. However, the drop was not as significant as with the charging
IV tests. For the extreme case of 50% reduced evaporator airflow, neither unit's capacity or EER dropped more than 25%. Return air leakage from hot attic spaces was simulated by assuming adiabatic mixing of the indoor air at normal conditions with the attic air at high temperatures. Effective capacity and EER both decreased with increased return air leakage. However, power consumption was relatively constant for all variables except outdoor temperature, which meant that for the same power consumption, the unit delivered much lower performance when there was return air leakage. The increase in sensible heat ratio (SHR) with increasing leakage showed perhaps the most detrimental effect of return air leakage on performance, which was the inability of the unit to absorb moisture from the environment.
V ACKNOWLEDGMENT I want to thank Dr. Dennis O'Neal for his help and guidance throughout this study. I also want to thank Dr. N.K. Anand and Dr. Mark Holtzapple for their suggestions and for serving on my committee. I would like to thank everybody at the Energy Systems Laboratory, especially Curtis, Frank, Kelly, Mike, and Pat. I would also like to thank my friend Joel Bain, who also worked long hours for the completion of this research project. Finally, I would like to thank my family: Angel F. Rodriguez, Maria Giraldez, Mariangelli Rodriguez, and Pedro J. Rodriguez. They have always been very supportive of my work and they are my best friends.
TABLE OF CONTENTS VI
TABLE OF CONTENTS (continued) VH
TABLE OF CONTENTS (continued) V1U
LIST OF TABLES IX
LIST OF TABLES (continued) X
LIST OF FIGURES XI
LIST OF FIGURES (Continued) Xll
LIST OF FIGURES (Continued) X1U
LIST OF FIGURES (Continued) XIV
LIST OF FIGURES (Continued) XV
LIST OF FIGURES (Continued) XVI
LIST OF FIGURES (Continued) XV11
1 CHAPTER I INTRODUCTION Heating and air conditioning systems account for approximately 20% of the total electrical energy use in the residential sector (HVAC Applications, ASHRAE, 1991). Electrical demand for electric utilities in the summer is usually greater when outdoor temperatures are higher. Two important contributors to this demand in residences are electric heat pumps and air conditioners. Both have lower capacity and efficiency with increasing outdoor temperatures. Because electric utilities are interested in reducing peak summer electrical demand, they have begun to focus attention on the performance of residential air conditioners at high outdoor temperatures. In recent years, many electric utilities have provided rebates to residential customers for purchasing high efficiency air conditioners and heat pumps. The rebates have helped increase the demand for higher efficiency air conditioning units. However, even the most efficient system will not perform as expected if it is not installed properly. Installation and maintenance items such as improper amount of charge in the system, reduced evaporator airflow, and air leakage in the return air duct from a hot attic space are very important in the determination of the performance of these units at high outdoor temperatures. The purpose of this research is to experimentally quantify the effect of these installation items on the high-temperature performance of air conditioners and heat pump systems. The test variables in this research are refrigerant charge, evaporator air flow rate, The format of this thesis conforms to that of the ASHRAE Transactions
2 and amount of return air leakage. The variables used to describe the performance of the air conditioners are: capacity, Energy Efficiency Ratio (EER), and power. The increased interest on the effect of these installation items on performance comes from the desire of electric utilities to reduce their peak demand on hot summer days. Of particular interest is whether high-efficiency air conditioners and heat pumps really perform better than older units at high outdoor temperatures. If a unit is not installed properly, it will not operate at its optimum level, even if it is a high-efficiency unit. There is a lack of data available on the effects of maintenance items on performance, especially on reduced evaporator airflow. More research is needed to evaluate how widespread these installation problems are, their overall effect on air conditioner performance, and the energy and peak demand penalties created by these effects. This project addresses the problem by providing much needed data and analysis of the effect of these maintenance items on high-temperature performance of air conditioners. This study is divided into ten chapters. In chapter II, a literature review was done to summarize what research had been conducted in this area. A detailed description of the experimental apparatus and procedure follows the literature review in Chapters III and IV. Chapters V and VI discuss the results for the refrigerant charge tests on two units. The next two chapters (VII and VIII) summarize the results for the reduced evaporator airflow tests. Chapter IX presents the return air leakage results, and the last chapter includes the conclusions, as well as recommendations for further work.
3 CHAPTER II LITERATURE REVIEW The performance of residential and commercial air conditioners at high outdoor temperatures is important to the electric utilities in the desert Southwest, where outdoor temperatures can reach 110 F (43.3 C) on hot summer days (ASHRAE Fundamentals, 1993). As outdoor temperatures increase, the thermal load on buildings increases while the efficiency and capacity of air conditioners decrease with increasing outdoor temperatures. This means that air conditioners have their worst performance when electrical demand is usually greatest. Table 2.1 - Climatic Conditions for Selected U.S. Cities* Design temperatures represent values that have been equaled or exceeded by 1%, 2.5%, and 5% of total hours during the months of June through September (Northern Hemisphere) Air conditioners and heat pumps are rated with a standard energy efficiency value, called the seasonal energy efficiency ratio (SEER). It is based on a set of tests specified in the Air Conditioning and Refrigeration Institute (ARI) test procedures (ARI, 1989). A higher value of SEER for an air conditioner should produce better performance over the whole cooling season. The performance of an air conditioner can also be affected by
4 several installation and maintenance factors. There is a need for quantitative data to properly analyze the effect of these factors on the high temperature performance of air conditioners. Air leakage into the return duct from a hot attic space reduces the capacity and efficiency of the unit. The reduced capacity causes the unit to run longer and use more energy. Under or overcharging a system can have a significant effect on performance. For example, an air conditioner with capillary tube expansion whose charge is set to produce optimum efficiency for 82 F (27.8 C) may not produce peak performance when the outdoor temperature is 100 F (37.8 C) (Farzad and O'Neal, 1993). This discussion of the relevant literature summarizes the data available on the effects of maintenance items on air conditioner performance. This review has been divided into the three maintenance items to be discussed in these project: improper charging, return air leakage, and reduced evaporator airflow. IMPROPER CHARGING OF AIR CONDITIONING UNITS Improper charging can be defined as either a refrigerant overcharge or undercharge when compared to the charge specified by the manufacturer. It can have a detrimental effect on the system, both in terms of equipment life expectancy as well as performance and efficiency. An undercharged condition creates a higher than normal superheat as well as lower subcooling. This can have an adverse effect on compressor motor winding cooling (Houcek and Thedford, 1984). Superheating is a term used to describe the sensible heating of the refrigerant vapor, at constant pressure, in the evaporator to a temperature above the saturation temperature corresponding to the evaporator pressure. The effect of superheat on capacity depends on the refrigerant used and other system conditions. Subcooling is a
5 term used to describe the cooling of the liquid refrigerant, at constant pressure, to a temperature below the saturation temperature corresponding to the condenser pressure. Subcooling of the liquid refrigerant increases the capacity of the system and decreases the amount of vapor produced in the expansion device during the expansion process. The long term effect can be premature compressor failure. Refrigerant undercharging also decreases the system capacity. Overcharging can reduce the life expectancy of a compressor due to the effects of liquid floodback, slugging, and motor overheating caused by abnormal compression ratios (Houcek and Thedford, 1984). Floodback is the continuous return of liquid refrigerant in the suction line to the compressor. Flooding causes oil dilution, which results in poor lubrication and overheating of bearing surfaces. Slugging is the occasional return of liquid refrigerant to the compressor. In extreme cases, slugging can produce broken components in the compressor. Because the density of the refrigerant vapor is greater, the compressor pumps against higher head pressures. It can be concluded that for overcharging, even though the cooling capacity could be about the same or even greater than for the correctly charged case, the efficiency of the system decreases because of the increased compressor outlet pressures. Several studies have been published on incorrect charging of air conditioners and heat pump systems. Houcek and Thedford (1984) conducted experiments using a lv 2 -ton split system. The dry bulb and wet bulb temperatures of the air entering the indoor coil were 80 F (26.7 C) and 67 F (19.4 C), respectively. In their research, several outdoor temperatures were used: 70 F (21.1 C), 82 F (27.8 C), 95 F (35 C), and 100 F (37.8 C). Their results showed that the undercharge condition affected the system operating efficiency and operating cost by causing a rapid decline in total capacity as ambient
6 temperature increased (Table 2.2). The capacity dropped from approximately 16000 Btu/h (4.69 kw) at 70 F (21.1 C) outdoor to 10250 Btu/h (3.00 kw) at 95 F (35 C) outdoor. The resulting drop in the Energy Efficiency Ratio (EER) was also very pronounced. At the same time, the electrical demand increased at a slower rate compared to the proper charge case as well as the overcharge case. Electrical demand at 95 F (35 C) ambient temperature was 1.82 kw at 23% undercharging compared to 1.98 kw at proper charge. Their results at 95 F (35 C) ambient temperature for overcharging on system performance show a slight increase in capacity over the baseline charge. That increase was from 17500 Btu/h (5.13 kw) at 23% overcharging to 16000 Btu/h (4.69 kw) at proper charge. For overcharging, the system showed a higher electrical demand than for normal charge. That increase was from 2.05 kw at 23% overcharging compared to 1.98 kw at proper charge. Table 2.2 - Performance Data at Different Charging Conditions* * At 95 F outdoor temperature. (Source: Houcek and Thedford, [1984]) Farzad (1990) studied the effect of improper charging for three expansion devices: a short-tube orifice, a thermostatic expansion valve, and a capillary tube. He found that the amount of charge in the system affected performance, but the amount of degradation of performance depended also on the type of expansion device used in the air conditioning unit. His results showed that units using thermostatic expansion valves (TXV) were less sensitive to charge amount than units with capillary tubes. In the -15% to +10% charge range, the capacity and EER of the TXV unit were relatively constant, while the capillary tube unit showed a definite decrease in capacity and EER when improperly charged,
Figure 2.2 - Normalized EER for Various Charging Conditions and Expansion Devices (Source: Farzad, [1990]) 7
8 (Figures 2.1 and 2.2). To obtain optimum capacity and efficiency, units with capillary tube expansion must be carefully charged to their proper amount. Orifice and TXV units will perform close to optimum if charged within 10% to 15% of the recommended charge level. Robinson studied the effect of charge on a three ton (10.5 kw) split-system heat pump operating in the cooling mode. The heat pump had a scroll compressor and utilized an orifice for the expansion device. He conducted tests for outdoor temperatures ranging from 82 F (27.8 C) to 105 F (40.6 C) and for charge levels ranging from 11 lbs (5.0 kg) to 14 lbs (6.4 kg) in one lb. (0.45 kg) increments. The charge needed to produce optimum performance of the unit was found to vary with outdoor temperature (Table 2.3). Robinson's results also show that a specific refrigerant flow rate produced maximum capacity at both 95 F (35 C) and 105 F (40.6 C) outdoor temperature. This means that if the optimum refrigerant flow rate was obtained at a specific outdoor temperature, it was not attainable at any other outdoor room temperature without changing the orifice size or charge level. Maintaining the optimum refrigerant flow rate throughout all outdoor temperature changes would yield better performance. This could be accomplished by changing the amount of refrigerant charge as the outdoor temperature changed, which would not be practical. Another way would be to change the size of the expansion device to maintain a specific flow rate. For example, instead of using an orifice, a TXV could be adjusted to maintain the superheat amount that corresponds to the optimum refrigerant flow rate. Table 2.3 - Optimum charge levels for different outdoor ambient temperatures* * (Source: Robinson, [1993])
9 RETURN AIR LEAKAGE Leakage in residential air distribution systems is likely to have a large impact on energy consumption, peak utility demands, and ventilation in a significant fraction of houses in the United States (Modera, 1989). In regions of the U.S. in which duct systems pass through unconditioned spaces, air infiltration rates into a residence will typically double when the distribution fan is turned on. Also, average annual air infiltration rate is increased by 30% to 70% due to the existence of the distribution system (Modera, 1989). According to Modera (1989), increases in peak electricity demands due to duct leakage can be as high as four kw per house in hot and humid climates, assuming that the return ducts pass through the attic. Increases in peak load per house on the order of one to two kw are likely in less extreme climates, or with less extreme return duct conditions. Using a simplified analysis, duct leakages in Sacramento, CA, and West Palm Beach, FL, were calculated to cause between two and ten MWh/yr increases in annual energy consumption. These results should be applicable to most of the Southern United States. It was observed that return-side leakage represented a surprisingly large fraction (64%) of total duct leakage. Their results showed that annual energy consumption was strongly dependent on the location of the return duct. Also, it was observed that a system with an attic return consumed between one to five MWh/yr more energy than a system with a crawlspace return. A research project in the Northwest compared whole house leakage and energy use of electric forced-air and room-heater-equipped homes (Lambert and Robison, 1989). They concluded that whole-house thermal losses of ducted homes are on average 40% greater than for unducted homes. Their results should not be surprising, because ducts would have to be perfectly insulated not to experience losses. However, the magnitude of
10 the observed effects was greater than expected. In their research (Lambert and Robison, 1989), there were four different groups of houses tested for air leakage. There were two control groups and two test groups. The control groups were houses built according to current regional practice. The test groups consisted of homes built under the highly energy-efficient Model Conservation Standard (MCS). This standard required substantially above-code insulation, infiltration reduction measures, heat recovery ventilation, and well insulated ducting. Both the control and test groups were then subdivided in two more groups, houses with or without forced-air heating systems. Their results showed that homes with ducted forced-air heating had more whole-house air leakage than homes without ducting. That air leakage difference averaged 26% for current practice homes. The incremental leakage and thermal losses due to the presence of ducting were substantially lower in highly energy-efficient homes (MCS). Incremental leakage and thermal losses for the MCS ducted group were 22% and 13%, respectively. Their results suggested that current construction practices associated with forcedair heating systems should be reviewed (Lambert and Robison, 1989). Substantially better performance was obtained by the ducted MCS test group. However, even the MCS test group had much greater leakage than the control homes (unducted). The authors agreed that leakage of the ductwork itself was a significant part of the problem, and that conduction losses from ducting must have played some role in the thermal losses. In a companion study of more than 20 ducted homes (Robison and Lambert, 1989), measured leakage of return ducts was found to be about twice the amount of supply ducts (Table 2.4). The predominance of return duct leakage over supply duct leakage confirmed
11 the investigators' theory that installers were more careful to seal seams on supply ducts (Robison and Lambert, 1989). Table 2.4 - Measured Leakage and Pefcent Increase in Total House Leakage Due to Return and Supply Ducts* * (Source: Robison and Lambert, [1989]) According to Cummings and Tooley (1989), return leakage of 15% attic air can reduce air-conditioner capacity by 50% or more during the peak cooling hours of the day, when maximum capacity is needed. Twenty percent return leaks of outdoor air can increase cooling season energy use by about 20%. If the source of the leak is attic air, the increase may be as much as 100%. They agreed with Robison and Lambert's research in that leaks in the return air ducts are typically larger than supply duct leaks. The importance of sealing supply leaks is obvious, because conditioned air is being lost. However, the problems of return leaks are not widely recognized. Energy savings from duct repair can be quite large (Cummins and Tooley, 1989). A 20% return leak coming from outdoor air at 95 F (35 C) dry-bulb temperature and 75 F (23.9 C) dew point temperature would reduce the net air conditioner capacity by about 30%. If this same return leak comes from an attic at 130 F (54.4 C) dry-bulb temperature and 85 F (29.4 C) dew point temperature, then the net air conditioner capacity may be reduced by 95%. The cost of repairing these systems is quite low. Based on estimates from a Florida company which specializes in these repairs, the cost for diagnosing the house and duct system should be about $50. The cost for the repair itself is typically less than $100 ($60
12 for a simple fix and as much as $150 for a more complex repair). From these results and from the fact that the cost of repairing these leaks is relatively low, it is evident why electric utilities are finding this an exceptional opportunity to reduce their peak demand at a very low cost. More research is needed to evaluate how widespread the problem of air leakage is, to determine ways of preventing and correcting return leaks, and to assess the energy and peak demand penalties created by these leaks. REDUCED EVAPORATOR AIRFLOW As evaporator air flow is reduced, cooling capacity and energy efficiency ratio (EER) decrease (Palanni, O'Neal, and Haberl 1992). For their tests, the cooling capacity decreased linearly until about 50% evaporator air flow, where it decreased much faster. For 90% air flow reduction, the cooling capacity decreased by 76%. Reductions in EER were similar to reductions in cooling capacity. The EER decreased linearly for decreasing evaporator airflow, then after 50% airflow, the decrease became non-linear and the drop was greater. The main conclusion was that to maintain sufficient cooling, at least 50% of the rated evaporator air flow was needed. Table 2.5 - Summary of Evaporator Airflow Results* (Source: Palanni, O'Neal, and Haberl, [1992]) Another investigator studied the effect of evaporator airflow on the frost/defrost performance of a heat pump (Payne and O'Neal, 1993). Low airflow degraded orifice performance by lowering the condensing temperature and reducing the inlet pressure. Low
13 airflow also lowered the system's heating capacity compared to the base case. The major effect of low airflow on system performance seemed to be the accelerated decrease in evaporating temperature. Low airflow lowered the evaporating temperature an average of 40% from base case conditions, which caused an acceleration in frost formation. These studies have demonstrated the detrimental effect on air conditioning performance caused by the installation items to be studied: improper charging of the system, reduced evaporator airflow, and leakage into return air ducts from hot attic spaces.
14 CHAPTER m EXPERIMENTAL APPARATUS The objective of the experiment was to study the effect of evaporator airflow, refrigerant charge, and return air leakage on the performance of residential air conditioners and heat pumps. The data collected included refrigerant and air mass flow rates, pressures and temperatures of refrigerant throughout the system, and power consumption of the unit. The experimental apparatus consisted of (1) the psychrometric rooms, (2) indoor and outdoor test sections, (3) the test air conditioners, and (4) installation and data acquisition. PSYCHROMETRIC ROOMS The air conditioning units were tested in the psychrometric rooms of the Energy Systems Laboratory at Texas A&M University. These rooms can control temperature and humidity for both indoor and outdoor sections of the unit. Dry-bulb and wet-bulb temperatures can be maintained within ±0.2 F during steady-state operation. The rooms were designed for testing systems with cooling capacities of up to 10 tons. Electric resistance heaters and chilled water coils are used to maintain room temperature. Reheat in each room was provided by four banks of 9.9 kw electrical strip heaters in the conditioning ductwork. The cooling coil was supplied with a chilled water ethylene glycol solution from a 150 ton chiller. A 1000 gallon chilled water thermal storage tank was mounted in the system to stabilize the chilled water temperature, while reducing the cycling of the chiller. Steam from a gas-fired boiler and dehumidification coils controlled the humidity in the rooms. The dehumidification coils were fed from the same circuit as the cooling coils. The boiler supplied steam directly into the supply air duct.
15 The indoor test section consisted of the indoor air flow chamber and the indoor coil corresponding to the specific test performed. The air flow chamber fan draws conditioned air from the indoor room through the indoor test section. An adjustable damper was used to maintain the amount of air flow specified for each unit. After leaving the air flow chamber, conditioned air was routed back into the indoor room. The outdoor room section is basically the condensing unit (compressor and outdoor coil). The conditioned outdoor air entered the outdoor coil and was exhausted by the unit fan back into the room. INDOOR AND OUTDOOR TEST SECTIONS The indoor test section consisted of the indoor air flow chamber, the test section, and the indoor fan coil unit. The placement of the indoor fan coil unit depended on the manufacturer's recommendation for the specific unit, which varied for the different tests (See Table 3.1 for a summary of the indoor and outdoor coils). Conditioned air from the indoor psychometric room passed through the indoor fan coil unit, insulated airflow ductwork, and the indoor airflow chamber. A 12-element thermocouple grid placed at the entrance of the fan coil unit measured entering air dry bulb temperature. Another 12- element thermocouple grid was used at the exit of the unit to provide exit dry-bulb and wet-bulb conditions. See Figure 3.1 for a sketch of the test section and the fan coil unit. A booster fan located at the exit of the flow chamber was used to draw air through the air duct. Air velocity through the test section was controlled by dampers located on the exit of the booster fan. The outdoor test section consisted of the outdoor fan coil unit. Air was drawn in through three sides of the outdoor unit and discharged through the top of the unit. A sampler mounted around the outdoor coil measured dry-bulb and dew-point temperatures.
Figure 3.1- Layout of the Psychrometric Rooms Showing the Placement of the Indoor and Outdoor Units (Numbers indicate measurement points found in Table 3.2) 16
17 TEST AIR CONDITIONERS For the refrigerant charging tests, two air conditioning units were used. One was a three ton (10.5 kw), split-system air conditioner with a scroll compressor and a short tube orifice as the expansion device. The other unit tested was a 3.5 ton (12.3 kw), split system air conditioner with a scroll compressor and a thermostatic expansion valve. For the reduced evaporator airflow test, there were also two units tested. One was a 3.5 ton (12.3 kw), split-system heat pump with a scroll compressor and a thermostatic expansion valve. The other unit was a 3.5 ton (12.3 kw), split system heat pump with a reciprocating compressor and a short tube orifice as the expansion device. For the return air leakage test, only one air conditioning unit was tested. It was a three ton (10.5 kw), split-system air conditioner with a scroll compressor and a thermostatic expansion valve. Table 3.1 has more detailed information on all the units tested. INSTRUMENTATION The instrumentation for all tests was divided into air-side, and refrigerant-side measurements. The air-side temperature measurements for both the inlet and outlet of the indoor coil unit were made using 12-element type-t thermocouple grids. Wet-bulb sensors were also used for both the inlet and outlet of indoor coil unit. For the outdoor unit, the only air-side temperature measured was the inlet air temperature. It was measured with a single type-t thermocouple located in the sampling duct surrounding the outdoor unit. The refrigerant-side measurements consisted of temperature and pressure measurements throughout the refrigerant lines. The refrigerant-side test points are shown on Figure 3.4 and the test points refer to the channels described in Table 3.2. With the exception of the refrigerant flow rate, the outdoor unit power, and the air flow differential pressure, the rest of the measurements were temperatures (dry-bulb, wet-bulb, dew point)
18 and pressures. A typical temperature probe used to obtain temperature on refrigerant lines is shown on Figure 3.3. These probes were 1/16 inch diameter and the total length of the probe inside the tube was at least 3/8 inch. The probes were inserted as close to the centerline of the copper tubing as possible. At each point that a temperature measurement was taken, pressure transducers were used to measure refrigerant pressure. A standard T-connection was put into the refrigerant line and a ball valve attached to one end. The pressure transducer was then attached to the valve. The main advantage of this arrangement was that it allowed the removal of the transducer without any losses in the refrigerant charge. Refrigerant mass flow was measured with a Coriolis-type mass flow meter. The flow meter was placed on the liquid line after the condensing unit (Figure 3.4). Table 3.1 - Description of the Test Air Conditioners and Heat Pumps* 1 (1) Refrigerant Charge, (2) Reduced Evaporator Airflow, (3) Return Air Leakage
Figure 3.3 - Typical Temperature Probe 19
Figure 3.4 - Data Acquisition Points in the Refrigerant Side of the System (Refer to Table 3.2 for more details) 20
21 DATA ACQUISITION The data acquisition system converted signals coming from all the sensors in the indoor and outdoor rooms into temperatures, pressures, flow rates, or power. A data logger was used to collect data from the testing apparatus. The logger was linked to a computer where the data were visually displayed during testing. Once a test was complete, the data were transferred to another computer for processing. A total of 22 channels were monitored during testing. Each channel was scanned by the logger at 30 second intervals. Since each test was twenty minutes long, a total of at least 40 time intervals were recorded for each test. A description of the data acquisition test points is listed in Table 3.2. Table 3.2 - Detail on the Data Acquisition Format
22 The parameters used to describe the performance of the air conditioners were capacity, energy efficiency ratio (EER), power consumption, and sensible heat ratio (SHR). Other values such as superheat, subcooling, various pressures, and air temperature drop across the evaporator were taken to better explain the trends in the performance curves. To calculate capacity, it was necessary to know the enthalpies leaving and entering the evaporator, as well as the air flow rate. The evaporator inlet and exit temperatures and pressures were used to determine the values for enthalpy. Refrigerant temperatures and pressures at the evaporator and expansion device were used to determine the values for superheat and subcooling respectively. The energy efficiency ratio (EER) was calculated from the capacity and power measurements.
23 CHAPTER IV EXPERIMENTAL PROCEDURE The purpose of the tests was to quantify the effect of installation problems on the high temperature performance of residential air conditioners. Each of the tests was steady state, with data averaged over a 20 minute interval. Even though ARI did not have performance tests for outdoor temperatures higher than 95 F (35 C), the measurements conformed to ARI tolerances for steady state cooling tests (ARI, 1989). The exception of this was the use of a 20 minute interval for data collection. ARI standards require one hour of data averaging for rating purposes. However, because of the large number of tests required, the length of tests were reduced to 20 minutes. The indoor conditions remained at 80 F (26.7 C) dry-bulb (DB) and 67 F (19.4 C) wet-bulb (WB) temperatures for all tests except for the return air leakage test. The outdoor temperatures ranged from 95 F (35 C) to 120 F (48.9 C). The experimental work included the three installation items in this study: improper charging of air conditioning units, reduced evaporator airflow, and return air leakage. Before discussing the individual test procedures, the installation of the unit is presented since it was the same regardless of the test performed. The first step was to install both the indoor and outdoor units. The next step was to set up all the sensors discussed in the last chapter and attach all of the refrigerant lines. A vacuum was pulled on the system for at least two hours to remove any moisture in the lines. Refrigerant was weighed and added to the system according to manufacturers' recommendations. The refrigerant used for all tests was R-22. The unit was charged when the indoor room was at 80 F (26.7 C) DB and 67 F (19.4 C) WB, and the outdoor room was at 95 F (35 C) DB
24 temperature. After the unit was charged, the airflow through the evaporator was set to its recommended setting by using a valve on the indoor air flow chamber. IMPROPER CHARGING OF AIR CONDITIONING UNITS Two split system air conditioners were used for this test. The first unit was a three ton (10.5 kw) capacity unit with a short tube orifice as the expansion device. The other unit was rated at 3.5 tons (12.3 kw) and had a thermostatic expansion valve (TXV) as the expansion device (Table 4.1). Table 4.1 - Description of the Units Tested for Improper Refrigerant Charging The three ton (10.5 kw) air conditioner was charged to a subcooling of 11 F (6.1 C), which was its recommended subcooling. There was no information on recommended subcooling for the 3.5 ton unit, but it had a factory charge of 11 lbs and 3 ounces (5.1 kg). Therefore, the unit was charged to a subcooling of 8.5 F (4.7 C). The refrigerant was weighed as it was added to the system. This amount of R-22 was designated as the baseline case. After the baseline tests were completed, the next step was to perform the overcharge and undercharge tests (Tables 4.2 and 4.3). The indoor section remained at 80 F (26.7 C) DB and 67 F (19.4 C) WB. For the orifice unit, the charge was increased
25 in increments of 10% for the overcharging tests up to 50% overcharging. For the TXV unit, the charge was increased in increments of 9% for the overcharging tests up to 27% overcharging. Overcharging can create conditions where liquid enters the compressor, particularly at high outdoor temperatures (Farzad, 1990). Therefore, for the orifice unit, the proposed tests only went to 20% overcharging for the 120 F (48.9 C), and 40% for the 110 F (43.3 C) outdoor temperature (Table 4.2). For the TXV unit, the proposed tests only went to 9% overcharging for the 120 F (48.9 C) (Table 4.3). Table 4.2 - Charging Tests for the Orifice unit* * Indoor room conditions: 80 F (26.7 C) DB, 67 F (19.4 C) WB Table 4.3 - Charging tests for the TXV unit* Indoor room conditions: 80 F (26.7 C) DB, 67 F (19.4 C) WB After the overcharge tests were completed, refrigerant was removed until the system was again charged to the baseline amount. Tests with indoor temperatures of 80 F
26 (26.7 C) DB and 67 F (19.4 C) WB, and 95 F (35 C) DB outdoor were then done. The data were compared to the original baseline data to verify that the system was at the same conditions as before. After the check was completed, the undercharge tests were started. Refrigerant was removed in 10% increments down to 30% undercharging for the orifice unit (Table 4.2). For the TXV unit, refrigerant was removed in 9% increments down to 36% undercharging (Table 4.3). Tests were run at 95 F (35 C), 110 F (43.3 C), and 120 F (48.9 C). The results for the refrigerant charge tests will be presented in graphical form and will quantify the degradation in performance due to over/undercharging of the system. Plots of capacity, power, EER, superheat, subcooling, and refrigerant flow rate will be presented as functions of both outdoor temperature and amount of charge. REDUCED EVAPORATOR AIRFLOW Two units were tested for reduced evaporator airflow. Both units were 3.5 ton (12.3 kw), split system heat pumps. One had TXV expansion and a scroll compressor while the other had orifice expansion and a reciprocating compressor (Table 4.4). Tests were run for outdoor temperatures ranging from 95 F (35 C) to 120 F (48.9 C). Table 4.5 shows a detailed description of the airflow test with all the outdoor temperatures. The first test was 100% airflow. The valve in the indoor airflow chamber was adjusted to reduce the airflow through the evaporator for the next tests. Table 4.4 - Description of the Units Tested for Reduced Evaporator Airflow
27 Table 4.5 - Reduced Evaporator Airflow Test as a Function of Outdoor Temperatures Indoor room conditions: 80 F (26.7 C) DB, 67 F (19.4 C) WB RETURN AIR LEAKAGE A 3.5 ton (12.3 kw) split-system air conditioner with a scroll compressor and TXV expansion was used for the return air leakage (Table 4.6). A series of tests were developed to evaluate the impact of a fixed amount of air leakage on system performance. The test leakage ranged from 5% to 25%, with attic conditions ranging from 130 F (54.4 C) to 150 F (65.6 C), and humidities ranging from 10% to 35%. The outdoor temperatures were between 100 F (37.8 C) and 120 F (48.9 C). The impact on effective capacity, EER, SHR, and power were quantified. Table 4.6 - Description of the Units Tested for Return Air Leakage The leakage from a hot attic was simulated by using two groups of tests. One group of tests was at the normal return air conditions of 75 F (23.9 C) DB and 50% relative humidity (RH) in order to obtain the indoor air properties and the performance of the unit with no leakage. For the other group of tests, evaporator entering air conditions
28 were adjusted based on the calculated adiabatic mixing of the normal return air conditions with the various attic conditions. For example, assuming that the attic conditions were 130 F (54.4 C) and 10% RH, with 15% mixing of this air with 85% of air at 75 F (23.9 C) DB and 50% RH would yield indoor room conditions of 82.6 F (28.1 C) DB and 65.0 F (18.3 C) WB (Appendix B). The assumption made here was that a hypothetical system at the new indoor room conditions was equal to the real system at the baseline indoor conditions with return air leakage. Two sets of data were used to make the calculations, one at the baseline indoor conditions of 75 F (23.9 C) DB and 50% RH, and another with the modified indoor conditions. To quantify performance of the air conditioner under simulated return air leakage conditions, variables such as effective capacity, effective EER, and effective SHR were defined. The equations were similar to the common capacity, EER, and SHR equations. However, the values for the evaporator entering conditions were the ones from the baseline test. The values for the evaporator exiting conditions were the ones from the test at the modified indoor conditions. After the effective capacity was calculated, the effective energy efficiency ratio (EER) was calculated by dividing the effective capacity by the power consumption values from the modified tests. The effective sensible heat ratio was calculated by dividing the effective sensible capacity by the effective total capacity. See Appendix B for a sample calculation of capacity, EER, and SHR.
Table 4.7 - Return Air Leakage Tests 29
30 CHAPTER V EXPERIMENTAL RESULTS FOR IMPROPER CHARGING OF AN A m CONDITIONER (ORIFICE EXPANSION) The unit tested for improper charging was a three ton (10.5 kw) split system air conditioner with short tube orifice expansion and a scroll compressor (Table 5.1). The amount of refrigerant charge in the system was varied to study its effect on the performance of the air conditioner. For each charge, the air conditioner was subjected to outdoor temperatures of 95 F (35 C), 110 F (43.3 C), and 120 F (48.9 C). The indoor conditions were set at 80 F (26.7 C) dry-bulb and 67 F (19.4 C) wet-bulb. Capacity, power, and energy efficiency ratio (EER) were used to measure the performance of the system. Refrigerant (R-22) was added until the manufacturer's recommended subcooling and superheat levels were obtained. That amount of refrigerant was recorded as the baseline charge. For this unit, the baseline charge amount was 6 lbs 12 oz (3.1 kg) (Table 5.1). Data were taken for the baseline charge at the three outdoor temperatures. Refrigerant was added in the increments needed for each of the overcharging tests. Refrigerant was then removed until the system was again at the baseline charge. Refrigerant was slowly removed and weighed until reading the desired charge for each of the undercharging tests. Table 5.1- Description of the Charging Test Unit #1
31 CAPACITY The capacity of the unit was based on the measurements on the air-side of the evaporator. The orifice unit was subjected to charge levels ranging from -30% charging to 50% overcharged. It was also subjected to outdoor temperatures ranging from 95 F (35 C) to 120 F (48.9 C). The capacity dropped as the charge decreased from full charge for all outdoor temperatures. The capacity showed a peak that was dependent on outdoor temperature. At 95 F (35 C), the peak capacity of the unit occured at 10% overcharging (See Figure 5.1). The capacity for this case was 34,638 Btu/h (10.1 kw) compared to 33,379 Btu/h (9.8 kw) at full charge. The highest capacity values at 110 F (43.3 C) and 120 F (48.9 C) outdoor temperatures also occured at 10% overcharging. However, the difference between the values at 10% overcharging and the baseline charge values were less than 1.4% for 110 F (43.3 C), and less than 0.5% for 120 F (48.9 C). At 20% or greater overcharging, the capacity started to decrease, but the drop was less than for undercharging conditions. For example, at 95 F (35 C) outdoor temperature, the drop in capacity for +20% charging from baseline charge is 0.7%, compared to an 24.3% drop in capacity for -20% charging from baseline charge. POWER CONSUMPTION AND ENERGY EFFICIENCY RATIO (EER) For this unit, the EER showed similar trends to the capacity curves. Figure 5.2 shows the EER as a function of outdoor temperature and refrigerant charge. For a given charge, as the outdoor temperature increased, the EER always decreased. The maximum EER occured at 10% overcharging and 95 F (35 C) outdoor temperature. This maximum value for EER was 10.74 Btu/Wh (3.15 kw/kw). Similar to the capacity curves, as the outdoor temperature increased, the peak of the EER curves shifted to the left, from +10%
32 charge at 95 F (35 C) to close to full charge for 120 F (48.9 C). The maximum EER was 8.61 Btu/Wh (2.52 kw/kw) for 110 F (43.3 C) outdoor temperature and +10% charge. At 120 F (48.9 C), the maximum EER value occured at full charge and its value was 7.38 Btu/Wh (2.16 kw/kw). The drop in EER was more noticeable for the low-charge conditions than for the high-charge conditions at any outdoor temperature. Power consumption increased with both increasing outdoor temperature and increasing refrigerant charge (Figure 5.3). For example, at 95 F (35 C) outdoor temperature, a change in charge from -30% to +50% produced an increase in power consumption from 2.84 kw to 3.31 kw. At 110 F (43.3 C) outdoor temperature, a change in charge from -30% to +40% produced an increase in power consumption from 3.31 kw to 3.79 kw. At 120 F (48.9 C) outdoor temperature, a change in charge from -20% to +20% produced an increase in power consumption from 3.84 kw to 4.09 kw. Figure 5.1 - The Short-Tube Orifice System Total Capacity for Various Outdoor Temperatures and Charging Conditions
33 Figure 5.2 - The Short-Tube Orifice System EER for Various Outdoor Temperatures and Charging Conditions Figure 5.3 - The Short-Tube Orifice System Power Consumption for Various Outdoor Temperatures and Charging Conditions
34 SUPERHEAT AND SUBCOOLING TEMPERATURES Measurement of superheat and subcooling temperatures requires measurement of both refrigerant temperature and pressure at the outlet of the condenser (subcooling) and outlet of the evaporator (superheat). The instructions that came with the air conditioning unit recommended specific values for superheating and subcooling temperatures. These values were used as a guide to charge the unit. Due to the combined errors associated with the temperature and pressure measurements, any superheat or subcooling temperatures between 0 F and 1 F (0.6 C) presented in this section were assumed to be saturated. The superheat and subcooling temperatures for the short-tube orifice unit were sensitive to charge. The superheat temperature for this orifice unit decreased for increasing outdoor temperatures (Figure 5.4). However, the superheat decreased much faster for increasing charge levels. For 95 F (35 C) outdoor temperature, the superheat decreased from 51.0 F (28.3 C) at -30% charging to saturated conditions at +50% charging. At the baseline charge case, the superheat decreased from 12.0 F (6.7 C) at 95 F (35 C) to saturated conditions at 120 F (48.9 C). The highest superheat temperature was obtained at 95 F (35 C) and -30% charging condition. A TXV unit would have kept a relatively constant superheat, therefore optimizing the refrigerant flow rate. However, refrigerant flow rate increased with both increasing charge and outdoor temperature (Figure 5.6). Subcooling increased with increasing charge for all the outdoor temperatures with the exception of 120 F (48.9 C), at -10% charging. For 95 F (35 C) outdoor temperature, the subcooling increased from 1.3 F (0.7 C) at -30% charging to 18.3 F (10.2 C) at +50% charging (Figure 5.5). In general, subcooling decreased with outdoor temperature. At the baseline charge case, the subcooling decreased from 10.9 F (6.1 C) at 95 F (35 C) to 7.9 F (4.4 C) at 120 F (48.9 C). The highest subcooling temperature of 18.3 F (10.2 C) was obtained at 95 F (35 C) and +50% charging.
35 Figure 5.4 - The Short-Tube Orifice System Superheat for Various Outdoor Temperatures and Charging Conditions Figure 5.5 - The Short-Tube Orifice System Subcooling for Various Outdoor Temperatures and Charging Conditions
36 Figure 5.6 - The Short-Tube Orifice System Refrigerant Flow Rate for Various Outdoor Temperatures and Charging Conditions SENSIBLE HEAT RATIO The sensible heat ratio (SHR) is defined as the ratio of the sensible capacity to the total system capacity. As SHR values increase, the system can provide the same capacity but a smaller fraction of the capacity its devoted to dehumidification. SHR decreased with increasing refrigerant charge (Figure 5.7). At 95 F (35 C) outdoor temperature, the sensible heat ratio ranged from 0.84 at -30% charging to 0.74 at +50% charging. At 110 F (43.3 C) outdoor temperature, the sensible heat ratio ranged from 0.82 at -30% charging to 0.78 at +40% charging. At 120 F (48.9 C) outdoor emperature, the sensible heat ratio ranged from 0.86 at -20% charging to 0.80 at +20% charging. The highest SHR value was 0.86 at 120 F (48.9 C) outdoor temperature and -20% charging. The lowest SHR value was 0.73 at 95 F (35 C) outdoor temperature and +20% charging. For most charging conditions, the SHR showed a small increase with increasing outdoor temperatures.
37 Figure 5.7 - The Short-Tube Orifice System Sensible Heat Ratio for Various Outdoor Temperatures and Charging Conditions PRESSURE DIFFERENTIAL AND CONDENSER PRESSURE As the outdoor temperature increased, the pressure differential from the inlet of the expansion device to the exit of the condenser increased (Figure 5.8). However, the variations of pressure differential with increasing charge were small. For example, at 95 F (35 C), the difference between the largest and smallest pressure differential was 4.3%. The largest pressure differential was 247.6 psi (1.7 MPa) at 120 F (48.9 C) and +20% charging. The condenser pressure increased with increasing charge, but started to level off from full charge. The condenser pressure also increased for increasing outdoor temperature (Figure 5.9). The highest condenser pressure was 348.29 psia (2.4 MPa) for +20% charge and at 120 F (48.9 C) outdoor temperature.
38 Figure 5.8 - The Short-Tube Orifice System Pressure Differential for Various Outdoor Temperatures and Charging Conditions Figure 5.9 - The Short-Tube Orifice System Condenser Pressure for Various Outdoor Temperatures and Charging Conditions
39 SUMMARY OF EXPERIMENTAL RESULTS Total capacity peaked at +10% charging for 95 F (35 C) outdoor temperature, between full charge and +10% for 110 F (43.3 C), and at full charge for 120 F (48.9 C). The highest capacity reading was at +10% charging and 95 F (35 C) outdoor temperature with a value of 34,638 Btu/h (10.1 kw). The trend for EER was very similar than for total capacity, with a peak at +10% charging for 95 F (35 C) outdoor temperature, between full charge and +10% for 110 F (43.3 C), and at full charge for 120 F (48.9 C). The highest EER reading was at +10% charging and 95 F (35 C) outdoor temperature with a value of 10.74 Btu/Wh (3.15 kw/kw). The capacity and EER dropped faster for the undercharged conditions than for the overcharged conditions. Both capacity and EER decreased with increasing outdoor temperature. Total power consumption increased linearly with both increasing charge and increasing outdoor temperature. Superheat decreased with increasing charge and with increasing outdoor temperature. Subcooling increased with increasing refrigerant charge, while it decreased with increasing outdoor temperature. Sensible heat ratio decreased with increasing charge, and showed a small increase with outdoor temperature.