Design & Engineering Services LOW-VOLTAGE TECHNOLOGY SCOPING STUDY: IMPACT OF SMALL AIR CONDITIONING UNITS ON ELECTRICAL SYSTEM VOLTAGE RECOVERY Demand Response Emerging Markets and Technologies Program Prepared by: Design & Engineering Services Customer Service Business Unit Southern California Edison April 15, 2006
Acknowledgements Southern California Edison s (SCE) Design & Engineering Services (D&ES) group is responsible for this project in collaboration with Tariff & Program Services (TP&S). It was developed as part of SCE s Demand Response, Emerging Markets and Technologies program under internal project number DR 05.05. D&ES s Refrigeration and Thermal Test Center (RTTC) project manager Rafik Sarhadian conducted this technology evaluation with overall management by Carlos Haiad of D&ES and Lauren Pemberton of TP&S. The D&ES team would also like to acknowledge the contributions and hard work of Richard Bravo of SCE s Transmission & Distribution, Loic Gaillac of SCE s Electric Vehicle Technology Center, and of Bruce Coburn and Ramin Faramarzi of RTTC. For more information on this project, email rafik.sarhadian@sce.com or carlos.haiad@sce.com Disclaimer This report was prepared by Southern California Edison (SCE) and funded by California utility customers under the auspices of the California Public Utilities Commission. This work was performed with reasonable care and in accordance with professional standards. However, neither SCE nor any entity performing the work pursuant to SCE s authority make any warranty or representation, expressed or implied, with regard to this report, the merchantability or fitness for a particular purpose of the results of the work, or any analyses, or conclusions contained in this report. The results reflected in the work are generally representative of operating conditions; however, the results in any other situation may vary depending upon particular operating conditions. Southern California Edison Page ii
Table of Contents Objective... 1 Background... 1 Smart Electric Meters... 2 Smart Thermostats... 2 Smart Breakers... 2 Smart Fuses... 2 Smart Direct Load Control Devices... 2 Smart Under-Voltage Protection Devices... 2 Air Conditioning Unit Selection and Preliminary Laboratory Testing... 3 Compressor Contactor Test... 5 Compressor Stall Test... 5 Real Power / Reactive Power Characteristic After Stall Condition Tests... 6 General Conclusions and Next Steps... 8 List of Figures Figure 1. Delayed Voltage Recovery 7/13/04 at 15:35 PST at SCE s Devers Substation.... 1 Figure 2. EVTC Outdoor Unit (Condenser) Controlled Environmental Room... 4 Figure 3. EVTC Indoor Air-Handling Indoor Environmental Room... 4 Figure 4. EVTC Grid-Simulator Power Conditioning/Sag Generator Unit... 4 Figure 5. EVTC Data Acquisition and Analysis Computers... 4 Figure 6. Incremental Voltage Drop Profile... 6 Figure 7. Stall Test of 3-ton Residential Air Conditioning Unit... 7 Figure 8. Schematic Voltage Recovery Profile... 7 Southern California Edison Page iii
OBJECTIVE This project focuses on: a) selecting and acquiring representative residential and small commercial air conditioning units common in Southern California Edison (SCE) service territory and b) performing preliminary laboratory testing on some of these units to provide initial insights on their electrical behavior under low-voltage conditions. This project is a component of a larger, multi-year research project (Evaluation of the Effects of Stalling Residential and Small Commercial Air conditioning Units on Electrical System Voltage Recover), which seeks to evaluate the impact of stalling air conditioning unit compressors on delayed electrical grid system voltage recovery after a system fault and to investigate practical solutions to reduce/eliminate the stalling problem. BACKGROUND Occasionally, the utility electric grid systems undergo voltage fluctuations caused by faults. These faults are caused by storms, accidents, debris on lines, and equipment failures. Normally, protective relaying is used to disconnect the problem lines and then voltage automatically returns to normal. There have been a few occurrences over the last several years where this voltage restoration has been delayed for up to 30 seconds. It is thought that air conditioning units are contributing to this delayed voltage restoration due to stalling compressors. When an air conditioning compressor stalls, its real and reactive power tends to increase which drags down system voltage. These stalled compressors do not disconnect from the electrical system until their internal thermal protection device is activated. Most large air conditioning systems have low-voltage protection; however, smaller units used in residential and light commercial buildings do not typically have this protection. Recent SCE events of delayed voltage recovery occurred on the Valley 500 kv system on three different occasions during the summer. These events were started by a fault on the 115 kv subtransmission system that cleared normally (4 to 6 cycles). After the faults were cleared, the voltage recovery was delayed for over 30 seconds (see Figure 1). Figure 1 shows the 500 kv voltage pattern as recorded at Devers Substation where the voltage dropped about 5%. The voltage at the 500 kv bus at Valley Substation dropped as much as 16% on the initial event as recorded by digital fault recorders. 4.7 2.4 0.2-2.1-4.3-6.6 15:35:43.00 15:36:13.00 15:36:43.00 15:37:13.00 15:37:43.00 15:38:13.00 15:38:43.00 Pacific Time Figure 1. Delayed Voltage Recovery 7/13/04 at 15:35 PST at SCE s Devers Substation. Southern California Edison Page 1
To understand this problem and its solutions, electrical system modeling will be performed, as part of the larger research project, to help predict system behavior under varying conditions. The current electrical system simulations do not properly account for the power characteristics of all air conditioning loads as an aggregate during low-voltage conditions and can yield inaccurate results. Results obtained from laboratory testing of air conditioning units under low-voltage conditions will enhance and improve the prediction reliability of electrical system simulation models. It is also important to investigate practical solutions to reduce or eliminate the air conditioning compressor stalling problem. These solutions may range from adding undervoltage protection to SCE installed direct load control (air conditioner cycling) devices to modifications/additions to the air conditioning units themselves. Currently, there are at least six possible ways/approaches to integrate/incorporate/implement mitigation technologies impacting the air conditioning unit: Smart electric meters Smart thermostats Smart breakers in electrical panels Smart fuses in air conditioning unit disconnects Smart direct load control (air conditioner cycling) devices Smart low-voltage protection devices SMART ELECTRIC METERS Evaluate the possibility of integrating low-voltage sensing and control capabilities into electric meters and use power line carrier or wireless communication to trip the breaker serving the air-conditioning equipment. SMART THERMOSTATS Evaluate the possibility of integrating low-voltage sensing and control capabilities into the thermostat serving the air-conditioning equipment. Under a pre-set low-voltage condition, the thermostat will reset the cooling temperature set point forcing the air-conditioning equipment to shut off. SMART BREAKERS Evaluate the possibility of integrating low-voltage sensing and control capabilities into the breaker(s) in the electric panel. Under a pre-set low-voltage condition, the breaker serving the air-conditioning equipment will trip. SMART FUSES Evaluate the possibility of integrating low-voltage sensing and control capabilities into the fuse in the disconnect box for the air-conditioning equipment. Under a pre-set low-voltage condition, the fuse serving the air-conditioning equipment would open. SMART DIRECT LOAD CONTROL DEVICES Evaluate the possibility of integrating low-voltage sensing and control capabilities into the direct load control device connected to the air-conditioning equipment. Under a pre-set low-voltage condition, the direct load control device will de-energize the air-conditioning equipment. SMART UNDER-VOLTAGE PROTECTION DEVICES Evaluate the possibility of integrating a more sophisticated low-voltage control capabilities into a low-voltage protection device connected to the air-conditioning equipment controls. Southern California Edison Page 2
Under a pre-set low-voltage condition, the low-voltage protection device will de-energize the air-conditioning equipment. These possible approaches need to be evaluated to determine the best solution in terms of cost, ease of installation, and time of implementation. Development of these solutions heavily depends on the findings of the laboratory testing of the residential and small commercial air conditioning units. The results of this large, multi-year research project will be used to improve load simulation models and devise methods to eliminate the delayed voltage recovery problem. These findings will potentially assist electric utilities in reducing the risk of local and/or systemwide voltage collapse. AIR CONDITIONING UNIT SELECTION AND PRELIMINARY LABORATORY TESTING A thorough investigation of the most common small air conditioning units found in SCE s service territory was conducted to determine the type and characteristics of the units to be tested. From this investigation, up to 10 air conditioning units were selected for purchased testing and purchased. Table 1 shows the initial list of the possible air conditioning units for purchase and testing. Table 1 Common Residential and Small Commercial Air Conditioning Units in SCE s Service Territory. Selection Unit Compressor Expension Efficiency Type Type Device Tonnage Vintage Make/Mfg. Condensor Unit Model No. Furnace/Coil Model No. Refrigerant Rating 1 Split Standard Scroll TXV 3 New X Lennox 10ACC-036-230 CK3BA036 R-22 SEER 10.25 X Carrier 38CKW036-3 CK3BA036 R-22 SEER 10.2 2 Split High Scroll TXV 3 New X Carrier 38BRG-036-30 CK3BA036 R-22 SEER 12.5 X Carrier 38TXA036-30 CK3BA036 R-410A SEER 13.0 X Lennox HSXA12-036 CK3BA036 R-410A SEER 12.9 X Bryant 597CN036-G CK3BA036 R-22 SEER 12.5 3 Split Standard Receip. TXV 3 New X Carrier 38CKC036-30 CK3BA036 R-22 SEER 10.5 4 Split High Receip. TXV 3 New X Trane 2TTB0036A1 (XB10) CK3BA036 R-22 SEER 12.0 5 Split Standard Scroll TXV 4 New X Bryant 561CE048-G CK3BA048 R-22 SEER 10.5 X Carrier 38TKB04837 CK3BA048 R-22 SEER 10.0 X Carrier 38TKW04832 CK3BA048 R-22 SEER 10.5 6 Split High Scroll TXV 4 New X Carrier 38BRC-048-3 CK3BA048 R-22 SEER 12.0 X Carrier 38TRA-048 CK3BA048 R-22 SEER 12.5 X Lennox HSXA12-048 CK3BA048 R-410A SEER 12.7 X Bryant 563C048A/D/E CK3BA048 R-22 SEER 12.5 X Bryant 597CN048-J CK3BA048 R-22 SEER 12.0 7 Split Standard Receip. TXV 4 New X Carrier 38CKG04836 CK3BA048 R-22 SEER 10.0 X Trane 2TTB0048A (XB10) CK3BA048 R-22 SEER 10.0 8 Split High Receip. TXV 4 New X Bryant CK3BA048 R-22 SEER 12.0 X Trane 2TTB2048A CK3BA048 R-22 SEER 12.0 9 Split Low TBD TXV 5 Old TBD 1989 Tempstar CA5060VKA1 TBD R-22 10 X Replacement unit Carrier 38TXA060-31 TBD R-410A SEER 13.0 10 Split Low TBD TXV 5 Old TBD 1982 Day & Night 568C8X0600000ASAA TBD R-22 9.5 X Replacement unit Carrier 38TXA060-31 TBD R-410A SEER 13.0 The tests were performed at SCE s Electric Vehicle Technology Center (EVTC) located in Pomona, California. The EVTC has a 3-phase, 50 kw grid-simulator power supply that can generate the voltage sags and delayed voltage recovery profiles necessary for the air conditioning unit testing. In addition, the EVTC has a temperature controlled chamber that was used to vary the temperature on the air conditioner outside unit (condenser) over the wide range representing summer conditions (85 to 130 F) in southern California. Figures 2 through 5 show the EVTC outdoor unit controlled environmental room, the indoor air handling indoor environmental room, the grid-simulator power conditioning/sag generator unit, and the data acquisition and analysis computers, respectively. Southern California Edison Page 3
Figure 2. EVTC Outdoor Unit (Condenser) Controlled Environmental Room Figure 4. EVTC Grid-Simulator Power Conditioning/Sag Generator Unit Figure 3. EVTC Indoor Air-Handling Indoor Environmental Room Figure 5. EVTC Data Acquisition and Analysis Computers Southern California Edison Page 4
Electrical performance characteristics of the air conditioning units were evaluated in three stages. The first test stage determined at what voltage the compressor contactor opens. If the contactor opens, the air conditioning unit shuts down and not stalls. This voltage determines the floor for the voltage in the second stage tests. This test is described in more detail in Section 3.1 Compressor Contactor Tests In the second test stage, tests were performed under a wide range of high ambient temperature conditions. The air conditioning units were set up in a controlled environment test chamber at EVTC. This stage determined at which voltage level and how fast the compressor stalls or electronic controls fail. If the air conditioning unit shuts down due to controls failing, it will be determined if the controls might attempt to restart the compressor when the voltage is still recovering. This test is described in more detail in Section 3.2 Compressor Stall Tests The third test stage will determine the real power (watt) and reactive power (VAR) characteristics of the compressors, after they stall, during delayed (30 seconds) voltage recovery. This test will also determine at what point the compressor thermal protection is activated. This test is described in more detail in Section 3.3 Real Power/Reactive Power After Stall Test. COMPRESSOR CONTACTOR TEST This test provided the voltage at which the compressor contactor opens (contactor dropout voltage). This test can be accomplished without the air conditioning compressor actually being in operation. The nominal voltage was imposed on the contactor and slowly reduced to see the point where the contactor opens. This voltage is important since it establishes the lowest voltage at which the compressor stall test needs to be executed. From prior tests by EPRI Solutions, this number is expected to be in the range of 50% to 60% of nominal voltage. COMPRESSOR STALL TEST This test evaluated the effects of step-by-step voltage drops on the performance of the air conditioning compressor and its controls. In particular, it ascertained if the lowering of the voltage caused compressor stalling or electronics failing. The step change in voltage took place in about 2 cycles (1/30 second) or less. If the unit did not stall in 30 seconds, the test was terminated and a new lower voltage test was initiated. An interval between each test allowed the compressor to return to its normal operating temperature. Tests were repeated for several voltage drops, but always starting at the nominal voltage. The tests started at 90% of nominal voltage and proceed in 5% voltage drop steps until the air conditioning compressor stalled instantly or the electronics fail (see Figure 6). If no stalling occurred, the test terminated when the voltage reaches the contactor dropout voltage. Because of the sensitivity of the compressor load to outdoor ambient temperature, this set of tests was repeated for several temperatures starting at 85 F and ending at 130 F in 5 F steps. A complete set of stall tests was performed for each temperature increment. During these tests, the indoor unit temperature was maintained at 80 F. For all the compressor stall tests, voltage, current, power and reactive power (all measured as root mean square (RMS) values) were recorded at a rate of at least 60 samples per second. In addition to the electrical measurements, temperatures were taken at various points on the outdoor unit (condenser) and the temperature, humidity, and flow of the air stream leading to the indoor unit. Refrigerant pressures were also measured on the suction and discharge of the compressor. Southern California Edison Page 5
240 V = 240-5% Voltage V = 240-10% V = 240-15% V = 240 - n% Time (seconds) Figure 6. Incremental Voltage Drop Profile These tests identified how fast the compressor stalls and at what voltage (compressor stall voltage). The findings from these tests are of key importance in determining the tripping function in the air conditioning unit direct load control (air conditioner cycling) switches or other protective devices. Figure 7 shows a sample of a stall test for a 3-ton residential air conditioning unit. Figure 7 shows the unit would operate without major problems when subjected to voltage sags down to 70% of nominal. However, at a voltage sag of 60%, the unit stalled with the normal load current of 14 amps increasing to 54 amps within a couple of cycles. This high load stayed on a line for almost 16 seconds before the unit was disconnected by its thermal overload protection. These tests were conducted with the outdoor unit (condenser) at a temperature of 85 F. REAL POWER / REACTIVE POWER CHARACTERISTIC AFTER STALL CONDITION TESTS This test recorded the voltage, current, real power (watt), and reactive power (VAR) of the air conditioning unit during a slow voltage recovery. In the cases observed in the Valley system, this recovery took place over 30 seconds. This test also determined when the compressor thermal protection was activated. The voltage started out at nominal voltage and then was lowered in a single step to the voltage that caused stalling for each outdoor unit test temperature (as determined from the second stage of tests Compressor Stall). After the unit stalled, the voltage was then gradually increased back to nominal voltage over a period of 30 seconds (see Figure 8). This voltage recovery profile was replicated with the grid simulator and reproduced the voltage recovery observed in the data from the Valley Substation events. Measurements of voltage, current, power and reactive power (all measured as RMS values) were made at a rate of at least 60 samples per second over the test period. In addition to the electrical measurements, temperatures were taken at various points on the outdoor unit (condenser) and the temperature, humidity, and flow of the air stream leading to the indoor unit. Southern California Edison Page 6
RMS Voltage 250 240 230 220 210 200 190 180 170 160 150 140 130 120 110 3-Ton A/C Unit Stalling Tests at EV Technology Center 8/18/2005 RMS Voltage RMS Current Approximately 16 Seconds 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 RMS Current 100 1 1485 2969 4453 5937 7421 8905 10389 11873 13357 14841 16325 17809 19293 20777 22261 23745 25229 26713 28197 29681 31165 0 Figure 7. Stall Test of 3-ton Residential Air Conditioning Unit Figure 8. Schematic Voltage Recovery Profile Refrigerant pressures were measured on the suction and discharge of the compressor. These measurements helped explain the shape of the voltage recovery curve observed in the Valley substation events. Southern California Edison Page 7
GENERAL CONCLUSIONS AND NEXT STEPS Preliminary laboratory testing performed at the EVTC provided initial insights on the electric behavior of residential and small commercial air conditioning units. The tests demonstrated that for low-voltage conditions below 70% of nominal, compressors on residential air conditioning units will go into a stall condition, which generates amperage five times greater than under normal operation conditions. Given these initial results, further tests will be conduced covering a range of small air conditioning unit characteristics under a wide range of high outdoor temperature conditions. Southern California Edison Page 8