by MJ Poort · Cited by 11 — Air Conditioning and Refrigeration Center. University of Illinois. Mechanical & Industrial Engineering Dept. 1206 West Green Street. Prepared as part of ACRC
64 pages
99 KB – 64 Pages
PAGE – 1 ============
University of Illinois at Urbana-Champaign Air Conditioning and Refrigeration Center A National Science Foundation/University Cooperative Research Center Applications and Control of Air Conditioning Systems Using Rapid Cycling to Modulate Capacity M. J. Poort and C. W. Bullard ACRC TR-238 May 2005 For additional information: Air Conditioning and Refrigeration Center University of Illinois Mechanical & Industrial Engineering Dept. 1206 West Green Street Prepared as part of ACRC Project #161 Urbana, IL 61801 Applications and Control of Systems Using Rapid-Cycling to Modulate Capacity (217) 333-3115 C. W. Bullard and P. S. Hrnjak, Principal Investigators
PAGE – 2 ============
The Air Conditioning and Refrigeration Center was founded in 1988 with a grant from the estate of Richard W. Kritzer, the founder of Peerless of America Inc. A State of Illinois Technology Challenge Grant helped build the laboratory facilities. The ACRC receives continuing support from the Richard W. Kritzer Endowment and the National Science Foundation. The following organizations have also become sponsors of the Center. Arçelik A. S. Behr GmbH and Co. Carrier Corporation Cerro Flow Products, Inc. Copeland Corporation Daikin Industries, Ltd. Danfoss A/S Delphi Thermal and Interior Embraco S. A. Ford Motor Company Fujitsu General Limited General Motors Corporation Hill PHOENIX Hydro Aluminum Adrian, Inc. Ingersoll-Rand/Climate Control Lennox International, Inc. LG Electronics, Inc. Manitowoc Ice, Inc. Modine Manufacturing Co. Novelis Global Technology Centre Parker Hannifin Corporation Peerless of America, Inc. Samsung Electronics Co., Ltd. Sanden Corporation Sanyo Electric Co., Ltd. Tecumseh Products Company Trane Visteon Automotive Systems Wieland-Werke, AG Wolverine Tube, Inc. For additional information: Air Conditioning & Refrigeration Center Mechanical & Industrial Engineering Dept. University of Illinois 1206 West Green Street Urbana, IL 61801 217 333 3115
PAGE – 3 ============
iiiAbstract Rapid cycling of the compressor can be used to modulate capacity as an alternative to variable speed a/c and refrigeration systems. This paper outlines design recommendations to optimize rapid cycling performance based on experimental results contrasting different heat exchangers and other components. Rapid cycling has inherent compressor lift penalties associated with larger mass flow rates relative to variable speed operation which need to be minimized. To design for optimal rapid cycling performance, it is also important to prevent dryout (superheating) in the evaporator during the off- cycle, a major penalty as cycles are lengthene d. By reducing the number of starts per hour, through increasing cycle lengths, compressor performance and reliability can be improved, and efficiency increased by reducing the number of startup power spikes. To increase cycle lengths while minimizing penalties, the off cycle performance in the evaporator should be optimized. During the off cycl e, two mechanisms contribute to off cycle cooling: additional boiling of refrigerant, and warming of the evaporator thermal mass. Experiments were done on a typical 2-ton residential system with a round tube and plate fin evaporator and compared to a similar 1-ton system with a microchannel evaporator to explor e the tradeoffs in evaporator design. Other components including a receiver an d a suction line heat exchanger were al so tested. Evaporator design should be focused on preventing dryout by increasing the refriger ant side area and preventing off cycle drainage. Since the condenser does not have the dryout problem, its only design changes would be to add thermal mass if economically feasible. For optimal performance under rapid cycling conditions, it was found that it would be best to combine flash gas bypass with a suction line heat exchanger to maximi ze performance during the off cycle and increase cycle lengths without incurring major penalties. This would allow the careful control of refrigerant boiling during the off cycle and maximize the thermal mass contribution to cooling the air during the off cycle.
PAGE – 4 ============
ivTable of Contents Page Abstract.. ..iii List of Fi guresv List of Tablesvi i Chapter1: Introduction..1 Chapter 2: Apparatus and methods..3 Chapter 3: Evaporator performance.4 Chapter 4: Evaporator design implications.8 Chapter 5: Condenser design11 Chapter 6: Other compone nts12 Chapter 7: Conclusions.15 References.16 Appendix A: Experimental Setup Ch anges17 Appendix B: Test matrix a nd data analys is method18 B.1 Test Matrix. ..18 B.2 Averaging Tair..18 B.3 Averaging Evaporator Metal Temp eratures18 Appendix C: Experimental Data20 C.1 Single Cycl e Analysis 20 C.2 Transients.. ..20 C.3 Off cycl e boiling. ..22 C.4 On Cycl e Recovery. 24 C.5 Transients in different capacity fractions 25 C.6 Comparison to conventional eva porator..29 C.7 Performan ce evalua tion.. 32 C.8 Pressure Lift.. .34 C.9 References.. .36 Appendix D: Data..37 Appendix E: Condenser behavior during rapid cycling.44 Appendix F: Water loops as thermal mass47 Appendix G: Components for low side desi gn53 G.1 High side receiver 53 G.2 Suction Line Heat Exchanger ..53 G.3 Low Side Receiver. .54 G.4 Flood Tank.. .55 G.5 Flash Gas Byp ass.. .55 G.6 Recommended C onfiguration.. ..56
PAGE – 5 ============
vList of Figures Page Figure 1.1 Diagram of temperature lif t for different ty pes of operation.. .1 Figure 3.1 Evaporator pressure drop 0. 4 runtime fraction, ~20 second cycle.. .4 Figure 3.2 Evaporator pressure drop 0. 4 runtime fraction, ~5 second cycle ..5 Figure 3.3 Evaporator instrume ntation ..5 Figure 3.4 Evaporator refrigerant, metal, and ai r temperatures for a 40 %, ~20 seco nd cycle ..6 Figure 3.5 Infrared picture sequence of the first two seconds of the front slab of the evaporator of a 20 second 40% cycle. .6 Figure 3.6 Evaporator refrigerant, metal, and ai r temperatures for a 40 %, ~5 second cycle. 7 Figure 4.1 Round tube and plate fin evapor ator temperatures, 10 sec 56% capacity.. ..8 Figure 4.2 Microchannel evaporator temper atures, 10.3 sec 0.58 capacity fraction.. .9 Figure 6.1 Diagram of the com ponent setup for rapid cycling.. .12 Figure B.3.1 Geometry of evaporator and the location of the numbered thermocouples..1 9 Figure C.2.1 Evaporator refrigerant pressure drop profiles of a 40%, ~5 second and ~20 second cycle..21 Figure C.2.2 Evaporator refrigerant, metal, and air temperatur es for a 40 %, ~2 0 second cy cle..22 Figure C.4.1 Infrared picture sequence of the first three seconds of the front slab of the evaporator of a 20 second 40% cycl e 25 Figure C.5.1 Evaporator refrigerant pressure drop profiles of an 80%, ~5 and ~20 second cycle26 Figure C.5.2 Evaporator refrigerant, metal, and air temperatur es for a 80 %, ~2 0 second cy cle..27 Figure C.5.3 Metal, refrigerant and air temper atures for 40% ~5 seco nd and VS operation2 8 Figure C.5.4 Metal, refrigerant and air temper atures for 80% ~5 seco nd and VS operation2 8 Figure C.6.1 Conventional evaporator pressu re drop, 10 sec 0.56 capacity fraction. ..30 Figure C.6.2 Microchannel evaporator pressu re drop, 10.3 sec 0.58 capacity fraction.. 30 Figure C.6.3 Conventional evaporator temp eratures, 10 sec 0.56 capacity fraction.. ..31 Figure C.6.4 Microchannel evaporator temperatures, 10 .3 sec 0.58 capacity fractio n (ignore noise spike).32 Figure C.7.1 Refrigerant si de temperature difference ..33 Figure C.7.2 Evidence of maldistribu tion from outlet ai r thermocouples .33 Figure C.8.1 Temperature lift at differen t variable speed ca pacity fractions 34 Figure C.8.2 Evaporator saturation temperatures for different cycle le ngths at 0.8 capacity fraction.35 Figure C.8.3 Temperature lift at different cap acity fractions35 Figure C.8.4 On cycl e temperat ure lift.. .36 Figure D.1 Evaporator refrigeran t, air and metal temperatures ..37 Figure D.2 Superheat at the evapor ator exit and comp ressor inlet 37 Figure D.3 Suction line heat exchanger ( bold), and nearby (thin) temperatures.. .38 Figure D.4 Pressure drop in the evaporator and condenser.. 38 Figure D.5 Refrigerant mass flow rate (at expa nsion valve) 39 Figure D.6 Compressor power use 39 Figure D.7 Sequence of IR camera pictures taken every second for the 20 seco nd cycle. 43 Figure E.1 Diagram of Condenser .44
PAGE – 6 ============
viFigure E.2 Condenser pressure drop in ~2 ton system. ..45 Figure E.3 Condenser pressure drop in ~1 ton system. ..45 Figure E.4 Condenser temperatures in 2 to n system (conventiona l 2 ton evaporator). 46 Figure E.5 Condenser temperatures on ~1 ton system (microchan nel ~1 ton system) 46 Figure F.1 Diagram of a se condary water loop with sh ell and tube condenser.. 48 Figure F.2 Diagram of a secondary water loop with storage tanks. ..50 Figure F.3 Approach temperatures of 0.3 KW water loop systems with the sa me heat exchangers.51 Figure G.1.1 Schematic and th ermodynamic cycle with a high side receiver. (Uribe). ..53 Figure G.2.1 Schematic and thermodynamic cycle with a suction line heat exchanger and a high side receiver. (Uribe) ..54 Figure G.3.1 Schematic and th ermodynamic cycle with a lo w side receiver. (Uribe).. ..54 Figure G.4.1 Schematic and thermodynami c cycle with a floo d tank. (Uribe) 55 Figure G.5.1 Schematic and thermodynamic cycle with a flash gas bypass. (Uribe) ..56
PAGE – 8 ============
1Chapter1: Introduction Air conditioning systems are usually designed to run at only one operating condition. In practice, however, energy can be saved by using variable speed systems to modulate the capacity. Rapid on-off cycling of a single- speed compressor could theoretically achieve results similar to variable speed systems, while avoiding the electrical cost and losses associated with inverter s. The goal of this project was to establish general guidelines for optimizing the overall design and compressor selection for rapid cycling systems. The internal design of the compressors is not addressed here, because the design of the rest of the system is independent of the compressor details. In rapid cycling the compressor is turned on & off in repeated cycles shorter th an ~70 seconds. Capacity modulation is regulated by co ntrolling the fraction of the cycle length du ring which the compressor is on. Since the cycles are short, the pressure lift oscillates around a mean th at is close to that of variable speed operation. Figure 1.1 shows schematically the saturation temperatures on the high and low side of the system and the temperature lifts for each operating condition. Figure 1.1 Diagram of temperature lift for different types of operation. Lower on-cycle temperature lift requires less compre ssor work for the same mass flow. In rapid cycling, when cycles are stretched out long enough, performance ap proaches that of conventional (long) cycling. On the other hand, when cycles are extremely s hort, performance approaches that of a variable speed system. For reasons of compressor reliability it is desirable for cycles to be stretc hed as long as possible without deviating too far from the temperature lift associated with variable speed operation. Experiments by Ilic et al. (2001), demonstrated th at on-off cycling of the compressor can achieve efficiencies within 2.5% of a variable speed system™s cy cle COP. These experiments also revealed that there was almost no off-cycle boiling so all the pressure drop and refrigerant-side temperature di fference occurred during the on cycle. For a 50% capacity case this meant that the refrigerant temperatur e difference and pressure drop were about twice as large duri ng the on cycle. During the off cycle, cooling continued as heat was transferred from the air Variable speed Rapid cycling (avg) ~20 min Conventional cycling ~10 sec cycle Time Temperature
PAGE – 9 ============
2to the thermal mass of the ev aporator, but off-cycle dryout (superheating) in sections of the evaporator degraded performance significantly during the long er cycle periods. Ilic identified the loss mechanisms unique to rapid cycling as the pressure drop, refrigerant temperature di fference and nonlinearity penalties. The pressure drop and refrigerant temperature differ ence loss terms stem from the need to pass tw ice as much refriger ant through the heat exchangers in half the time. The nonlin earity penalty refers to the differences between the on-cycle and whole-cycle average temperatures of the heat exch angers, these asymmetrical oscillations are especially noticeable in heat exchangers having low thermal capacitance. Experiments also indicated that it was be best to isolate the high and low side of the system using two solenoid valves to maintain the saturation pressures on both sides of the system during the off cycle. Ilic et al concentrated primarily on lengthening cycles by increasing the thermal capacitance of the heat exchangers in order to achieve temperature oscillations that remained nearly symmetrical over longer cycle periods, thus minimizing degradation of temperature lift associated with the exponen tial nature of heat transfer. The performance margin (2.5 % of variable speed system COP) is smaller than the in verter losses that would further degrade variable speed system COP. This paper describes subsequent experiments designed to explore the potential for further optimization of the overall system. Due to the high cost of increasing the evaporator and condenser thermal mass, and the difficulty of avoiding evaporator dryout during longer off cycles, a different approach was taken, involving installation of a microchannel evaporator in the same facility used by Ilic. Due to the greater refrigerant side area; an appropriately designed microchannel evaporator would have a smaller refrigerant-side temperature difference, and a smaller pressure drop due to the large number of multi-port tube s. Thus minimizing these variables and their associated temperature lift penalties in the variable-speed baseline case makes the evaporator less vulnerable to the losses incurred when these parameters increase during rapid cycling. Unfortunately the best-available prototype microchannel heat exchanger was far from an ideal evapor ator prototype because it had been designed as a CO 2 gas cooler and therefore had ports that we re too small and too little in number for our R22 system. Nevertheless it did prove adequate for testing hypotheses about the nature and magnitude of the loss mechanisms inherent in rapid cycling systems. Since the evaporator was still vulnerable to dryout, experiments remained focused on shorter cycle periods (5-20 seconds). Experimental results are summarized below, and followed by overall system design recommendations based on the comparison with results obtained using conventional heat exchangers and different arrangement of other components. For full details see the appendices.
PAGE – 10 ============
3Chapter 2: Apparatus and methods The experimental setup consisted of a residential split system installed in an instrumented outdoor and indoor chamber. Previous work was done on heat exchange rs from a two-ton, R-22 unitary rooftop air conditioning system (Trane TCH024100A) with a two-ton hermetic scroll compressor (Copeland model ZR22K3-TF5) as described by Ilic et al. The TXV was replaced by an el ectronic expansion valve which allowed more precise control while avoiding the inherent lag of a TXV which is problematic without a steady mass flow. The microchannel evaporator, suction line heat exchanger and 1 ton reciprocating compressor (Copeland model KAN 0075) were later installed by Uribe et al. (2003). Other changes made to the experimental setup primarily concerned the location of measurements and are further detailed in appendix A. All temperatures were measured by T-type thermoco uples and probes that were calibrated initially between an ice bath and room temperature, and again prior to each set of experiments by adjusting tare values at room temperature. The uncertainty of thermocouple measurements was estimated to be ± 0.1°C. To measure air flow rate, a 0-1fl water differential air pressure transducer was used to detect the pressure drop acr oss the nozzle with an error of ± 0.4 % of full scale. The absolute pressures were meas ured at the evaporator and compressor outlets with ±0.5 psia error. Pressure differences were also detected across both heat exchangers with ±0.125 psid error. Refrigerant mass flow was measured using a Coriolis type mass flow me ter within ±0.15% error. The power to the compressor, blower and heaters were measured using watt transducers with ±40 W error. Data from each experiment was obtained after the rapid-cycling system reached an equilibrium operating state characterized by st able repeatable cycles. All calculations were performed using 200-second av erages of data recorded at 0.2 second intervals. The data plots for single cycles show a typical cycle selected from the 200-second record.
PAGE – 11 ============
4Chapter 3: Evaporator performance Evaporator temperatures and pressure were recorded during steady rep eatable oscillations at rapid cycling conditions that could be compared to variable speed oper ation at the same cooling capacity. As cycle periods got shorter, the amplitude of oscillations diminished until the mean values of all thermodynamic variables approached those of variable speed operation. Refrigerant pressure drop across the evaporator was one of these oscillating measurements that displayed behavior important to understanding rapid cycling. While the magnitude of this pressure drop would have been lower with an optimally designed evaporator, its behavior yielded qualitative insights. Significant on-cycle pressure drop was not reached for about two seconds while the liquid inventory was being re-establis hed during the on cycle, and it continued to rise as the saturation temperature dec lined and more boiling occurred in the evaporator. After compressor shut down 8 seconds into the cycle, the pressure drop reveals that boiling continued for about 4 more seconds as the remaining liquid evaporated. Figure 3.1 s hows the whole-cycle pressure drop to be significantly higher than that measured during variable speed operation. Figure 3.1 Evaporator pressure drop 0.4 runtime fraction, ~20 second cycle. In shorter cycles such as the 5 se cond cycle in Figure 3.2, the pressu re drop did not have a chance to establish itself during the on cycle. It is also likely that the walls never dried out since there is not enough time to boil all the liquid from the evaporator. Due to the short cycl e length, the whole-cycle average pressure drop is much closer to that of variable speed opera tion, even though the compressor was only on during the first two seconds of the cycle. Thermodynamic performance of even shorter cycles would approach that of variable speed systems more closely. 010203040 506002468101214161820 Time [sec] Pressure [kPa] PP avg P vs on off
99 KB – 64 Pages