Cold Climate Heat Pump using Tandem Vapor-Injection Compressors

Transcription

1 Cold Climate Heat Pump using Tandem Vapor-Injection Compressors Bo Shen *[ ], Omar Abdelaziz, Van Baxter, Edward Vineyard Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831, U.S.A. *corresponding Author: Tel.: ; Abstract. Conventional air-source heat pumps (ASHPs) experience rather poor performance in cold climate areas. The heating capacity and efficiency of conventional ASHPs decrease significantly as the outdoor temperature decreases. The major R&D challenges are to limit this ASHP heating capacity and efficiency degradation at low and extremely low ambient temperatures. Vapor injection (VI) compressors are able to provide better efficiency and larger capacity at low ambient temperatures. A prototype air-source cold climate heat pump (CCHP), using tandem vapor injection (VI) compressors and inter-stage flash tank, was developed. The CCHP has two identical VI compressors in parallel, which works with a two-stage indoor blower and two-stage thermostat. At moderately low ambient temperatures, only one compressor is called, and at extremely low ambient temperatures, both the compressors are used. The prototype was installed in Fairbanks, Alaska and underwent field testing for six months. The CCHP successfully operated down to -30 F (-35 C) and was able to meet the building heating load with good efficiency in a wide range of ambient temperatures. At -30 F (-35 C), the CCHP delivered 75% heat pump capacity, relative to the capacity at 47 F (8.3 C), and the heat pump COP was 1.8. This paper will introduce the CCHP development and field testing results. Keywords: Vapor Injection Compressor, Tandem Compressors, Cold Climate Heat Pump. 1 Introduction As described by Khowailed et al. [2], in the U. S., the primary target market for cold climate heat pumps (CCHP) is the 2.6 million U.S. homes using electric furnaces This manuscript has been authored by UT-Battelle, LLC under Contract No. DE- AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (

2 2 and conventional air-source heat pumps (ASHP) in the cold/very cold region, with an annual energy consumption of 0.16 quadrillion (0.17 EJ). A high performance airsource CCHP would result in significant savings over current technologies (greater than 60% compared to electric resistance heating). It can result in an annual primary energy savings of 0.1 quadrillion ( EJ) when fully deployed, which is equivalent to 5.9 million tons (5.35 million MT) of annual CO2 emissions reduction. In cold climate areas with limited access to natural gas, conventional electric ASHPs or electric resistance furnaces can be used to provide heating. During very cold periods, the ASHPs tend to use almost as much energy as the electric furnaces due to their severe capacity loss and efficiency degradation. Presently, technical and economic barriers limit market penetration of heat pumps in cold climates. R&D efforts should be employed to overcome these barriers and develop high performance CCHPs that minimize, or even eliminate, the need for backup strip heating. A typical single-speed ASHP doesn t work well under cold outdoor temperature conditions typical of cold climate locations for three major reasons: 1. Too high discharge temperature: low suction pressures and high pressure ratios at low ambient temperatures cause significantly high compressor discharge temperatures, in excess of the maximum limit for many current compressors on the market. Furthermore, system charge of a heat pump is usually optimized in cooling mode, which leads to overcharge conditions in heating mode, further increasing the discharge temperature. 2. Insufficient heating capacity: heating capacity decreases with ambient temperature. The heating capacity at -13 F (-25 C) typically decreases to 20% to 40% of the rated heating capacity at 47 F (8.3 C) (~equivalent to the rated cooling capacity at 95 F (35 C)). As such, a single-speed ASHP, sized to match the building cooling load, is not able to provide adequate heating capacity to match the building heating load at low ambient temperatures, and supplemental resistance heat has to be used. 3. Low COP: heating COP degrades significantly at low ambient temperatures, due to the elevated temperature difference between the source side and demand side. For the CCHP development, cost-effective solutions should be identified to tackle these three issues. US Department of Energy (DOE) has set stringent performance targets for CCHPs as follows: 1) maintain at least 75% of the rated space heating/cooling capacity at -13 F (-25 C), relative to the rated heating capacity at 47 F (8.3 C), and 2) have a rated heating COP at 47 F (8.3 C) greater than 4.0. The 75% capacity criterion would result in a heat pump capacity approximately equal to the building heating load for a well-insulated home at -13 F (-25 C) in US climate Region V, for example, Minnesota (assumed to be the DHRmin load condition as defined by AHRI Standard 210/240 [1] for Region V), where the building heating load at -13 F (-25 C) is 80% of the building cooling design load at 95 F (35 C) ambient temperature. Researchers have investigated several cycle configurations for CCHP. Wang et al. (2009) [4] studied advanced vapor injection (VI) cycles. A VI cycle uses a vapor injection compressor. In a VI system, liquid from the condenser is expanded to a middle stage between the condensing and evaporating pressures, after phase separation,

3 3 the vapor is injected to the compressor injection port, and the liquid is further expanded and goes to the evaporator. VI cycles can be classified into two fundamental configurations: (a) Flash tank cycle and (b) Economizing heat exchanger cycle. Figure 1 shows the schematics of a VI cycle for each configuration. In a VI cycle with flash tank, two-phase refrigerant is separated into saturated liquid and vapor by a flash tank after the first expansion. It has the advantage of feeding 100% of saturated vapor to the compressor injection port. The two-stage cycle with economizing heat exchanger allows part of the liquid refrigerant at the condenser outlet to pass through an expansion valve before entering the economizer HX to further subcool the mainstream refrigerant coming from the condenser. The superheated intermediate pressure refrigerant leaving the economizer HX enters the intermediate compressor port. As a result, the separation with economizer HX will never be 100% as compared to the flash tank separation due to the limited surface area involved. The refrigerant flow rate and pressure entering the intermediate compressor port can be easily controlled using thermostatic expansion valves. The VI cycles are able to reduce compressor discharge temperature effectively by directly injecting low enthalpy refrigerant vapor into the compressor compression cylinder. They also increase the evaporating and heating capacities due to the lower enthalpy liquid refrigerant entering the evaporator after the interstage phase separation. Condenser Condenser Flash Tank Scroll Multi-Stage Compressor Economizer HX Scroll Multi-Stage Compressor Evaporator Evaporator Fig. 1. Vapor injection compressor coupled with a flash tank or economizer. Vapor injection (VI) compressors are able to provide better efficiency and larger capacity at low ambient temperatures. The economizing heat exchanger is usually more expensive than the flash tank, but easier to control. The flash tank can lead to better performance with a lower cost, but it is hard to control in a wide range of ambient temperatures, which may cause vapor underfeeding or overfeeding to the VI compressor. Although VI compressors shows superior performance under low ambient temperatures, a heat pump using a single-speed VI compressor still can t meet the 75% capacity goal at -25 C, as indicated by Shen et al. [3]. Shen et al. [3] discussed the development of a cost-effective CCHP, using two equal, single-speed compressors (tandem). Note that the system is relatively simple and comparable to conventional ASHPs with the exception of having two compressors in parallel, thus it is considered to be relatively more cost-effective than more

4 4 complex, variable-speed design approaches. Two options of tandem compressors were studied; the first employs two identical, single-speed compressors, and the second employs two identical, vapor-injection compressors. The investigations were based on system modeling and laboratory evaluation. Both designs have successfully met the performance criteria. Laboratory evaluation showed that the tandem, singlespeed compressor ASHP system is able to achieve heating COP = 4.2 at 47 F (8.3 C), COP = 2.9 at 17 F (-8.3 C), and 76% rated capacity and COP = 1.9 at -13 F (-25 C). This yields a HSPF = 11.0 (heating seasonal performance factor, per AHRI 210/240). The tandem, vapor-injection ASHP is able to reach heating COP = 4.4 at 47 F (8.3 C), COP = 3.1 at 17 F (-8.3 C), and 88% rated capacity and COP = 2.0 at -13 F (-25 C). This yields a HSPF = This paper introduces a field investigation for the second design, i.e. the tandem, vapor-injection ASHP. 2 Field Study 2.1 System Diagram For the field testing prototype, ORNL researchers applied an innovative configuration that allows for improved capacity modulation based on equal-size tandem VI scroll compressors. VI compressors are less prone to capacity degradation with decreasing the ambient temperature and the compressor discharge temperature is lower than non-vi compressors, as discussed earlier. The proposed control algorithm facilitates efficient operation over a wide range of ambient conditions. Figure 2 below illustrates the hardware system configuration for the field investigation. Indoor unit Fan Condenser Ma Ta Ta P T TXV P T Filter/Dryer Liquid Receiver 5 6 EXV 3 4 P P T Tandem VI compressors 4-Way Valve T Tamb Outdoor unit One-way check valve 1 2 Fixed orifice Flash Tank Injection line accumulator Suction line accumulator P T Evaporator Fan Fixed orifice Compressor Compartment T Fig. 2. Refrigeration cycle schematic: air-source CCHP using tandem VI compressors, interstage flash tank, and optimum inter-stage pressure control, field testing prototype. The tandem VI compressors are coupled with an inter-stage flash tank to separate phases and feed saturated vapor to the compressors VI port.

5 5 A suction line accumulator, an injection line accumulator, a liquid receiver and the inter-stage flash tank are used as charge buffers. The liquid receiver is connected to four one-way check valves (3, 4, 5, and 6) to maintain consistent inlet and outlet lines for both heating and cooling modes. The suction line accumulator and the injection line accumulator protect the compressors from liquid refrigerant entering. Two fixed-orifices are installed upstream and downstream of the flash tank to provide throttling, and control the suction and injection pressures for heating mode. An electronic expansion valve (EXV) is installed upstream of the liquid receiver to optimize the injection pressure as a function of the ambient temperature and compressor staging. Two check valves (1, 2) are installed to control the flow direction for cooling, heating and defrosting modes. A separate TXV is used for cooling mode only. A four-way valve is used to change the refrigerant flow direction as a typical heat pump does. A two-speed indoor blower is used. The EXV upstream of the liquid receiver was embedded with a static control function, to adjust the EXV opening as a function of the ambient temperature and compressor staging. That led to optimized head pressure and discharge temperature control as static targets, and prevented underfeeding or overfeeding refrigerant to the VI port. The fixed orifice before the flash tank differentiated the pressures of the liquid receiver and the flask tank, and the liquid level in the receiver can be varied by the EXV than going to the flash tank. 2.2 Field Testing Results We collaborated with the Cold Climate Housing Research Center (CCHRC) in Fairbanks, Alaska, USA, and used their lab space to host the field investigation. This site doesn t necessarily represent a typical residential or commercial building but was readily available and allowed the CCHP to experience a very harsh winter. Heating season operation started in September During the field test, the CCHP successfully ran down to -30 F (-35 C), with heating capacity >75% of the rated capacity and heating COP > 1.8. We developed a data acquisition (DAQ) system for the field data monitoring and control implementation. A DAQ controller and measurement devices from National Instruments Inc. were used to monitor the field operations. These were connected to an indoor host computer to process and store the data. The field testing data were sent back through Dropbox via internet connectivity. Figure 2 describes the system diagram and instrumentation, where P represents refrigerant pressure transducers; T represents refrigerant-side temperature measurements (all inserted probe thermocouples, except the one soldered on the evaporator exit tube surface); Ta means air side temperature measurements; Tamb means an ambient temperature measurement; Ma means the indoor air flow measurement using an air flow monitor having an array of pitot tubes. The air temperatures in and out of the indoor air handler were measured using T- type thermocouples. Three thermocouples were evenly placed at the entrance of the

6 6 indoor unit to measure the average return air temperature. At the exit of the indoor blower, three thermocouples were used to monitor the supply air state. Four pressure transducers were used to measure the refrigerant pressures entering and leaving the indoor coil, as well as entering and leaving the compressors. Another pressure transducer was used to measure the pressure entering the injection port. Four watt transducers were used to measure the powers of the outdoor fan, indoor blower and two compressors, individually. The DAQ system scanned all the sensors and recorded the data every half minute. Figure 3 illustrates the run time fraction of the second compressor to the total heat pump run time, changing with the ambient temperature. The second compressor operated more frequently as ambient temperature fell below 5 F. At -30 F (-35 C), the second compressor was still cycling at a 90% fraction, indicating that the prototype CCHP system still had reserve heating capacity and was able to meet the heating load of the lab space even at this extreme cold condition % 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% "-34.4 to -31.7C" "-31.7 to -28.9C" "-28.9 to -26.1C" "-26.1 to -23.3C" "-23.3 to -20.6C" "-20.6 to -17.8C" "-17.8 to -15C" "-15 to -12.2C" "-12.2 to -9.4C" "-9.4 to -6.7C" "-6.7 to -3.9C" "-3.9 to -1.1C" "-1.1 to 1.7C" "1.7 to 4.4C" "4.4 to 7.2C" "7.2 to 10C" "10 to 12.8C" "12.8 to 15.6C" Fig. 3. The second compressor run fraction, relative to the total run time versus ambient temperature. Figure 4 shows the field-measured average delivered heating capacities for each outdoor temperature bin for both single and dual-compressor operation. At -30 F (- 34 C) with two compressors, the average delivered capacity was 27,000 Btu/h (7.9 kw). This is 75% of the rated starting point capacity of the heat pump unit at 47 F (8.9 C) or 36,000 Btu/h (10.6 kw). Note that the average heating capacity per temperature bin decreases as the ambient temperature drops to ~5 F (-15 C) (with only one compressor operating for majority of the time). As the outdoor temperature decreased from 5 F to -30 F (-30 C), the second compressor began running for an increasing portion of the time allowing the CCHP system to increase its delivered heating capacity and meet building load down to the lowest temperature bin experienced in the field test. Figure 5 shows the average measured temperature rise in the indoor coil air stream across the indoor blower unit. The system room thermostat was set at 68 F (20 C), which caused the zone temperature to typically change from 66 F to 70 F (18.9 C to

7 C) over each heat pump run cycle. The indoor air flow rate was allowed to change with the compressor staging (High/Low). The measured low-stage indoor air flow rate was around 1200 CFM and the high indoor air flow rate was around 1650 CFM. Because the field testing lab space has a lower load profile than a typical building, only a single-speed compressor was normally needed above 5 F (-15 C). At moderate ambient temperatures, e.g. above 25 F (-3.9 F), the heat pump cycled for most time. All these factors led to an approximately uniform temperature rise from the return to supply air around 15 R (8.3 K). Delivered Capacity [W] Delivered Capacities "-34.4 to -31.7C" "-31.7 to -28.9C" "-28.9 to -26.1C" "-26.1 to -23.3C" "-23.3 to -20.6C" "-20.6 to -17.8C" "-17.8 to -15C" "-15 to -12.2C" "-12.2 to -9.4C" "-9.4 to -6.7C" "-6.7 to -3.9C" "-3.9 to -1.1C" "-1.1 to 1.7C" "1.7 to 4.4C" "4.4 to 7.2C" "7.2 to 10C" "10 to 12.8C" "12.8 to 15.6C" 1_Comp 2_Comp Total Fig. 4. Delivered heat capacities of running one or two compressors. Temperature Rise [R] Fig. 5. Average temperature rise from return to supply. Figure 6 presents the defrost cycle run time fraction relative to the total heat pump operation time. It can be seen that the defrosting frequency was minimal for the CCHP for two reasons: 1. During the outdoor temperature range most conducive to outdoor coil frosting, only one compressor was operating most of the time and frost formation was slow. This was due to the outdoor coil being relatively oversized compared to the single compressor capacity leading to higher evaporating temperatures than typical for conventional ASHPs.

8 8 2. When two compressors were needed at low ambient temperatures, the outdoor humidity level was very low and hardly any moisture condensed on the outdoor coil. Because the EXV opening was controlled as a function of the ambient temperature, the defrost frequency appeared to peak in the F (-12.2 to -9.4 C) outdoor temperature bin. 6% 5% 4% 3% 2% 1% 0% Fig. 6. Defrost time ratio relative to capacity delivered in each bin. Figure 7 illustrates average field-measured bin COPs for both single-and dualcompressor operation. The total COP (or overall system average bin COP) was calculated by the total energy delivery divided by the total energy consumed, including loss effects due to cycling, frosting/defrosting, and switching between one or two compressor operation. The single- and two-compressor COP curves do not include the defrost loss effect. During the defrosting operation, the heat pump always ran two compressors in the reversed cycle. It can be seen for the 45 F to 50 F (7.2 C to 10 C) bin the total COP is 4.0. This is about 12% lower than that obtained in earlier lab measurements for steady-state operation because of the cyclic loss effects. It is encouraging to see that, at -30 F (-35 C), the total COP was about 1.8, i.e. 80% more efficient than resistance heating. The field COP at -13 F (-25 C) is around 2.0. It is also interesting to note that for ambient temperatures below about -15 F (-26 C), single compressor operation appears less efficient than dual compressor operation. 5 COP [W/W] _Comp 2_Comp Total Fig. 7. Field COPs in heating mode.

9 9 3 Two Lessons Learned A low-pressure compressor protection shutoff control prevented system operation below about -30 F (-34 C) ambient due to a minimum suction pressure set point > 30 psia (207 kpa). However, the compressors could have continued to operate at even lower temperatures. The compressor discharge temperature was around 230 F at the -30 F condition, far below the maximum discharge temperature limit of 280 F (138 C). A lower suction pressure protection setting would allow the CCHP to operate down to perhaps even -40 F (-40 C). The field testing site chosen had a low building heating load profile, and the CCHP was sized to match the peak building load around -30 F (-34 C). This meant that two-compressor operation typically did not occur until the ambient temperature fell to around 5 F (-15 C). The CCHP was able to provide enough heating capacity, but the indoor air flow rates appear too high to deliver comfortable supply air temperatures. A redesign of the indoor blower or reconfiguration of the blower speed settings to reduce the air flow rate would have boosted the supply air temperature and the indoor comfort level. However; this would come at some system efficiency penalty due to higher condensing temperature in heating operation. 4 Conclusions CCHPs are targeted to the climate zones having a significant portion of heating energy delivered below 17 F (-8.3 C) for homes without connection to low cost gas supply. To develop a CCHP working under extremely low ambient temperatures, the first key is to make sure that the compressor(s) can operate without exceeding the compressor discharge temperature and pressure ratio limits. To accomplish this, optimizing the system charge for heating mode, properly sizing the HXs, and controlling discharge temperature are necessary. In addition, the compressor(s) should be able to tolerate high discharge temperature, for example, as high as 250 to 280 F (121.1 to C). To maintain a good ASHP efficiency at low ambient temperatures, it is most important to provide sufficient heat pump capacity and eliminate most of the resistance backup heating. At low ambient temperatures, heating load increases substantially; however, the heating capacity of a typical single-speed ASHP decreases steadily with lowered ambient temperatures, which degrades to 20% to 40% of the rated heating capacity. Thus, oversizing is mandatory for a CCHP, i.e. using only part of the compressor capacity to determine the rated heating capacity at 47 F (8.3 C) and meet the building design cooling load. On the other hand, the system should also provide good part-load efficiency in heating mode at moderate ambient temperature, as well as in cooling mode. These requirements point to using tandem compressors or variablespeed compressors with capacity modulation capability, among which the option of tandem compressors is more cost-effective. We developed lab and field testing prototypes and successfully met the US DOE s performance targets for CCHPs 1) to maintain heating capacity at -13 F (-25 C)

10 10 greater than 75% of the rated heating capacity at 47 F (8.3 C), and 2) heating COP at 47 F (8.3 C) greater than 4.0. We developed a lab prototype using tandem vapor injection compressors. The prototype reached 4.4 COP at 47 F (8.3 C); 88% heating capacity and 2.0 COP at -13 F (-25 C), and 3.1 COP at 17 F (-8.3 C), having a HSPF of Its seasonal performance is uniformly 5% higher than the using just tandem, single-speed compressors. We fabricated a prototype air-source CCHP system using equal size tandem VI compressors, coupled with an inter-stage flash tank, which was fully instrumented and embedded with a complete control algorithm including defrosting, compressor and fan staging, EXV modulation to control optimum inter-stage compressor as a function of the ambient temperature. The prototype was installed in Fairbanks, Alaska for field testing for the period of 09/2016 to 03/2017. The CCHP successfully operated down to -30 F (-34 C) and met the building heating load with good efficiency over a wide range of ambient temperatures. At -30 F (-34 C), the CCHP delivered 75% heat pump capacity, relative to the capacity at 47 F, and the heat pump COP was 1.8. Acknowledgements The authors thank Mr. Antonio Bouza, Technology Development Manager for HVAC, WH, and Appliances, Emerging Technologies Program, Buildings Technology Office at the U.S. Department of Energy for supporting this research project. References 1. ANSI/AHRI, Standard 210/240 with Addenda 1 and 2, 2008 Standard for Performance Rating of Unitary Air-Conditioning & Air-Source Heat Pump Equipment, Air Conditioning, Heating, and Refrigeration Institute, Arlington, VA, USA[2] 2. Khowailed, G., K. Sikes, and O. A. Abdelaziz., Preliminary Market Assessment for Cold Climate Heat Pumps, ORNL/TM-2011/422, Oak Ridge National Laboratory, August. 3. Shen, B., Omar Abdelaziz, Keith Rice, Van Baxter and Hung Pham, Cold Climate Heat Pumps Using Tandem Compressor, Conference Paper in 2016 ASHRAE Winter Conference, Orlando, FL. 4. Wang, X., Hwang, Y. and Radermacher, R Two-stage heat pump system with vapor-injected scroll compressor using R410A as a refrigerant, International Journal of Refrigeration, Vol. 32, pp