Zero Net Energy Retrofit Projects The Domes Final Performance Assessment & Verification ET Project Number: ET12PGE1442 Project Manager: Peter Turnbull Pacific Gas and Electric Company Prepared By: NORESCO 2540 Frontier Avenue, Suite 100 Boulder, Colorado 80301 Contact: Vaughn Engler Engineer I, Energy Services 303.459.7451 vengler@archenergy.com Issued: August 25, 2015 Copyright, 2015, Pacific Gas and Electric Company. All rights reserved.
ACKNOWLEDGEMENTS Pacific Gas and Electric Company s Emerging Technologies Program is responsible for this project. It was developed as part of Pacific Gas and Electric Company s Emerging Technology Technology Assessments program under internal project number ET12PGE1442. NORESCO conducted this technology evaluation for Pacific Gas and Electric Company with overall guidance and management from Anna LaRue at Resource Refocus LLC and Peter Turnbull at Pacific Gas and Electric Company. For more information on this project, contact pwt1@pge.com. LEGAL NOTICE This report was prepared for Pacific Gas and Electric Company for use by its employees and agents. Neither Pacific Gas and Electric Company nor any of its employees and agents: (1) makes any written or oral warranty, expressed or implied, including, but not limited to those concerning merchantability or fitness for a particular purpose; (2) assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, process, method, or policy contained herein; or (3) represents that its use would not infringe any privately owned rights, including, but not limited to, patents, trademarks, or copyrights. 1
ABBREVIATIONS AND ACRONYMS COP DHW EER GSHP HC HE HPWH HVAC PG&E PTAC PTHP ZNE Coefficient of Performance Domestic Hot Water Energy Efficiency Ratio Ground Source Heat Pump Heating Capacity Heat of Extraction Heat Pump Water Heater Heating, Ventilation, Air Conditioning Pacific Gas and Electric Company Packaged Terminal Air Conditioner Packaged Terminal Heat Pump Zero Net Energy 2
FIGURES Figure 1. Panorama of The Domes at Baggins End.... 7 Figure 2. Dome Exterior.... 7 Figure 3. Energy End-Use Breakdown.... 8 Figure 4. Dome Monitoring Points Schematic... 10 Figure 5. Inside a Typical Installation... 11 Figure 6. Ground Loop Temperature Sensor... 11 Figure 7. Dome 13 GSHP Heating System Temperatures... 15 Figure 8. Dome 13 GSHP Heating Performance... 15 Figure 9. Dome 14 GSHP Heating System Temperatures... 17 Figure 10. Dome 14 GSHP Heating Performance... 17 Figure 11. Dome 15 PTHP Heating System Temperatures... 19 Figure 12. Dome 15 PTHP Heating Performance... 19 Figure 13. Dome 13 GSHP Cooling System Temperatures... 21 Figure 14. Dome 13 GSHP Cooling Performance... 21 Figure 15. Dome 14 GSHP Cooling System Temperatures... 23 Figure 16. Dome 14 GSHP Cooling Performance... 23 Figure 17. Dome 15 PTHP Cooling System Temperatures... 25 Figure 18. Dome 15 PTHP Cooling Performance... 25 Figure 19. Actual Energy By End-Use (March 2014 February 2015) 29 Figure 20. Average Indoor Temperatures... 30 Figure 21. Dome Indoor Temperature Comparison... 31 Figure 22. Domestic Hot Water Comparison... 32 Figure 23. DHW Operation October... 33 TABLES Table 1. Summary of Baseline and Retrofit Domes.... 6 Table 2. The Domes Monitoring Points... 12 Table 3. Heating Energy (kwh) Use Comparison... 26 Table 4. Cooling Energy (kwh) Use Comparison... 26 Table 5. Plugs/Lights Energy (kwh) Use Comparison... 27 Table 6. Range Energy (kwh) Use Comparison... 27 Table 7. DHW Energy (kwh) Use Comparison... 28 Table 8. Total Energy (kwh) Use Comparison... 28 3
EQUATIONS Equation 1. Coefficient of Performance (COP)... 13 Equation 2. Heating Capacity (HC)... 13 Equation 3. Heat of Extraction (HE)... 13 Equation 4. Energy Efficiency Rating (EER)... 14 CONTENTS FIGURES 2 TABLES 3 EQUATIONS 4 CONTENTS 4 ABBREVIATIONS AND ACRONYMS 2 EXECUTIVE SUMMARY 5 1. INTRODUCTION 6 2. Domes Retrofit Process... 8 2.1 Energy Analysis... 8 2.2 Retrofit Technologies... 9 3. Energy Performance Verification... 9 3.1 Monitoring Equipment... 9 3.2 Energy Performance Monitoring Period... 11 3.3 Monitoring Plan... 11 4. Data Analysis and Results... 13 4.1 Heating Performance Equations... 13 4.2 Cooling Performance Equations... 13 4.3 Heating Performance Results and Discussion... 14 4.4 Cooling Performance Results and Discussion... 20 4.5 Overall Energy Use Comparison... 26 4.6 Domestic Hot Water Analysis... 31 5. Conclusion... 34 Appendix A.... 35 Drawings and Product Information... 35 4
EXECUTIVE SUMMARY In 2012, the University of California, Davis (UC Davis), Energy Institute initiated a project funded by the Emerging Technologies Program at the Pacific Gas and Electric Company (PG&E) to retrofit existing residential structures through energy efficiency improvements and with ground source heating and cooling systems to achieve zero net energy (ZNE) performance goals. The project team hoped to identify scalable, deployment- ready technologies that could hasten the transition to zero net energy homes in PG&E territory. In addition, it was a project goal that the technologies selected would be both leading- edge as well as readily available, reliable, and proven. This is exemplified in the selection of technologies such as LED lights, heat pumps, and basic envelope improvements. NORESCO, formerly Architectural Energy Corporation, was part of the project team to inform the choices of retrofit equipment, accurately estimate the potential energy savings, and later monitor the energy performance of installed systems. Initially, computer generated whole- house energy simulations were generated and compared to the metered utility information for the site. Analysis of the energy model showed that the greatest opportunities for energy savings would be to reduce the heating and cooling energy used, which led to the selection of the specific technologies. Four out of the fourteen residences at The Domes project site have been retrofitted with the selected technologies. One of these four domes did not receive the HVAC retrofit and is a partial baseline. Energy monitoring equipment has been installed at these four retrofitted Domes along with two other baseline Domes that did not receive any retrofit to improved envelope, lighting, shared DHW system, and heating/cooling equipment. The performance of the retrofitted Domes was ultimately compared to the performance of the three baseline Domes. NORESCO has collected a full twelve months of monitored performance data, spanning from February 2014- February 2015, which has provided insight into the performance of the heat pump technologies. It has also highlighted a number of potential performance issues. The heat pumps have proven to be effective at heating and cooling the retrofitted Domes and all three units are operating efficiently. However, the shared domestic hot water system is using significantly more energy than the existing baseline domestic water heaters and is causing the overall energy use of the retrofitted Domes to be much higher than the baseline. Some of the more significant findings are listed below. The heat pump water heater serving Dome 8 operates very efficiently, using less than half the energy as the other baseline Domes. The shared domestic hot water heat pump serving Domes 13-15 appears to have a failed compressor or some other issue causing it to operate very inefficiently (using electric resistance only). While there may still be potential for this system to run efficiently, this installation has demonstrated that special care and attention should be paid to heat pump water heaters. Occupant behavior has the greatest effect on HVAC and plug energy use and should be considered as an area of focus for further studies. The efficiency of the two ground source heat pumps does appear to be slightly improved over the packaged terminal heat pump, providing cooling around 10% more efficiently, however installation was significantly more costly and involved. The heat pumps are experimental proto- type units and may have room for improvement. 5
1. INTRODUCTION The Domes site in Davis, CA, consists of 14 detached Dome structure homes originally constructed in the 1970s. The Domes are located near the intersection of Orchard Park Circle and Orchard Park Drive near the UC Davis campus and are occupied by a diverse group of UC Davis students. Each Dome houses two students, and these residents generally embrace a sense of community and share resources such as a garden, some livestock, and frequently make use of The Commons area and the community Yurt. These structures are served by electricity only, from PG&E, and are all connected to one master site meter. The homes each have an approximate 450 ft2 floor plan on concrete slabs, with fiberglass and polyurethane wall construction, double- pane aluminum- framed windows, and wooden doors. The overall R- value of the envelope, including walls, doors, and windows, is between 20-23 (ft2 F Hour/BTU). The equipment in the homes prior to any retrofits includes refrigerators, electric ovens & ranges, electric domestic hot water heaters, CFL lighting, and electric space heaters. No cooling was in place. As part of this project, Domes 8 and 13-15 were retrofitted with slab insulation, new doors, skylights, and LED lighting. Domes 13-15 were also connected to a new central air- to- water heat pump water heater. The electric space heaters were removed and replaced with ground source heat pumps (GSHP) for Domes 13 and 14 and a packaged terminal heat pump (PTHP) for Dome 15. In both these cases, the replacement equipment provides cooling as well as heating. In addition to the envelope and lighting improvements, Dome 8 was retrofitted with a standalone air source heat pump water heater. The electric space heater was not removed and no cooling was added. NORESCO installed comprehensive monitoring equipment on site in order to quantify the energy performance of the various retrofits in Domes 8, 13-5 compared to Domes 6 and 10, which were not retrofitted. Table 1 provides of a summary of the retrofit and baseline domes. TABLE 1. SUMMARY OF BASELINE AND RETROFIT DOMES. RETROFITS DOME # SLAB INSULATION, DOORS, SKYLIGHTS, LED LIGHTING CENTRAL AIR-TO- WATER HEAT PUMP WATER HEATER STANDALONE AIR SOURCE HEAT PUMP WATER HEATER GSHP FOR HEATING AND COOLING PTHP FOR HEATING AND COOLING 6 (Baseline) 8 (Partial Baseline) X X 10 (Baseline) 13 (Retrofitted) X X X 14 (Retrofitted) X X X 15 (Retrofitted) X X X 6
FIGURE 1. PANORAMA OF THE DOMES AT BAGGINS END. FIGURE 2. DOME EXTERIOR. 7
2. DOMES RETROFIT PROCESS The verification and assessment of the retrofit technologies installed in the Domes was completed by NORESCO and UC Davis at the following: n The Domes 3 retrofitted homes 1 partial retrofitted/partial baseline home 2 baseline homes 2.1 ENERGY ANALYSIS At the beginning of the project, in order to inform the choices of retrofit equipment and accurately estimate the potential energy savings, computer generated whole- house energy simulations were created by NORESCO using REM/Rate and compared to the metered utility information. Model inputs were taken from drawings where possible, otherwise assumptions were made and adjusted to reflect historical energy bills. Note that there is only one meter for the entire Domes site, so calibration of the energy model is approximate. These models estimated the as- built total annual load energy performance of the homes based on the heating and cooling requirements for Sacramento, CA (Climate zone 3B). The estimated annual energy use for each Dome is approximately 6,000 kwh. The results of the energy models indicated that there was a substantial opportunity to improve the performance of the Domes during the heating season by addressing the thermal efficiency of multiple building components. Insulating the slab floor, replacing the windows & skylights, and improving the thermal properties of the door should all contribute to reduced energy consumption. Although the existing baseline Domes do not have mechanical cooling, in order to make a more accurate comparison with the retrofitted Domes (which do have cooling), a theoretical baseline Dome with cooling was modeled. During the cooling season, the largest controllable opportunity to save energy was determined to be replacing the windows and skylights, to reduce solar heat gain, and to increase the overall insulation of the Dome. FIGURE 3. E NERGY END-USE BREAKDOWN.
2.2 RETROFIT TECHNOLOGIES Based on the energy analysis results, technologies were selected and implemented in four of the Domes, but only three of these received HVAC retrofits. Details of the specific envelope, lighting, space conditioning, and water heating technologies installed are described in the following sections. 2.2.1 ENVELOPE AND LIGHTING IMPROVEMENTS During the summer of 2013, Domes 8 and 13-15 were retrofitted with R- 3.2 polyisocyanurate insulation around the concrete slab edge, new insulated entry doors, and fixed skylights in place of exhaust fans. LED lighting was also installed in the kitchen to replace the existing CFLs in each of the retrofitted Domes. 2.2.2 DHW IMPROVEMENTS An air- to- water AO Smith hybrid electric heat pump water heater was installed in a newly constructed enclosure adjacent to Dome 14 to serve the three retrofitted Domes 13 through 15. This system has a recirculating loop which maintains domestic hot water temperatures for Domes 13 and 15. The recirculating loop runs underground in order to reach Domes 13 and 15. Because it is immediately adjacent to the water heater, the hot water supply to Dome 14 is piped separately and is not part of the recirculation loop. A separate 50 gallon Geospring air- to- water heat pump was installed adjacent to Dome 8 and serves that Dome only with no recirculation loop. 2.2.3 HVAC IMPROVEMENTS An air- to- air AMANA PTH123 packaged terminal heat pump was installed in Dome 15 for heating and cooling. This same model of heat pump was modified by Enertech to function as a ground source water- to- air heat pump and installed in Domes 13 and 14. For more information, see the report on ZNE Retrofits at UC Davis. 1 3. ENERGY PERFORMANCE VERIFICATION The assessment methods used at The Domes included sub- metered end- use monitoring, monitoring of space temperature, outside air temperature, flow rates and temperatures of ground loop water, and subsequent analysis of the energy performance of the systems. The verification of the actual energy savings of the implemented retrofits was accomplished by comparing the sub- metered residences to the baseline residences. The baseline Domes (6 and 10) are similar to the retrofitted Domes but did not receive retrofit technologies. Dome 8 is a partial baseline which received the evnvelope and lighting retrofits and a new water heater but no retrofits to the HVAC system. The baseline Domes were sub- metered in the same fashion as the retrofitted Domes. 3.1 MONITORING EQUIPMENT The monitoring equipment utilized was installed by NORESCO and was in place from February 7, 2014 through March 31, 2015. Arrangements have been made with UC Davis and PG&E to keep the equipment in place beyond the end of this study. Equipment and data collection devices include the following equipment: 1 UC Davis Energy Institute, 2014. Zero Net Energy Retrofits at UC Davis. http://www.etccca.com/reports/zero-net-energy-retrofits-uc-davis
n Hobo U30 Remote Monitoring Stations Combination temperature/relative humidity sensors for indoor environmental monitoring Split- core current transformers for measuring circuit current Wattnode power meters for total energy use, PTHP energy use, and water heater energy use Surface- mounted temperature sensors for hot water and ground loop supply and return temperatures Seametrics flow meters for measuring ground water and hot water flow The Hobo U30 station serves as a data collection hub for the various sensors in the Dome. Using cell data transmissions, the Hobo U30 uploads all of its data daily to the hobolink website where it can be accessed remotely with a username and password. FIGURE 4. DOME MONITORING POINTS SCHEMATIC Figure 4 above shows approximate locations of the various sensors within the Domes. All of the Split- core current transformers and Wattnode power meters are located in or immediately adjacent to the electrical panel. This schematic applies to Domes 13 and 14; for Dome 15 the difference is that there is no ground loop and instead a temperature sensor for discharge air from the PTHP.
FIGURE 5. INSIDE A TYPICAL INSTALLATION FIGURE 6. G ROUND L OOP T EMPERATURE S ENSOR (INSULATION REMOVED TO SHOW MOUNTING) 3.2 ENERGY PERFORMANCE MONITORING PERIOD The scope of services performed by NORESCO began in December 2013 and continued through the end of February 2015. Due to changes in scope, baseline Dome monitoring equipment was not installed until February 7, 2014. During this monitoring period, there were quarterly summary reports which highlighted significant findings, data collection or analysis issues, and presented opportunities for additional energy savings. 3.3 MONITORING PLAN Table 2 outlines the monitoring points that were used to accomplish the verification tasks. General metering was performed at the panel level, where the loads were classified as either HVAC, range, lighting/plug loads (including appliances such as refrigerator and lamps etc.), and water heating. The retrofits that were identified for monitoring were presented categorically which is not necessarily representative of the field monitoring setup. For example, while the lighting was an individual retrofit, lighting is not isolated on its own separate circuit and thus lighting and plug loads are monitored together as one load.
3.3.1 THE DOMES MONITORING POINTS TABLE 2. THE DOMES MONITORING POINTS MONITORED ATTRIBUTE TREND POINT TREND INTERVAL UNITS DOME 6 (BASE) DOME 10 (BASE) DOME 8 (PARTIAL BASE) DOME 13 (RETRO) DOME 14 (RETRO) DOME 15 (RETRO) Total Dome Electric Main service power (Wattnode) 5 min kwh x x x x x x Lighting/ Plugs/ Appliances Lighting/Plugs circuit current 5 min Amps x x x x x x GSHP (Dome 13, 14 Heat/Cool) Total unit power (Wattnode) Heat pump ground water flow Ground water supply temperature 5 min kwh 5 min Gallons 5 min Degrees F x x Ground water return temperature 5 min Degrees F PTHP HP (Dome 15 Heat/Cool) Total power (Wattnode) Supply air temperature 5 min kwh 5 min Degrees F x Space Heater (Baseline Domes) Total current 5 min Amps x x x Envelope Indoor air temperature 5 min Degrees F Indoor relative humidity Outdoor air temperature 5 min % 5 min Degrees F x x x x x x Outdoor relative humidity 5 min % Range Range current 5 min Amps x x x x x x Heat Pump Water Heater Total power (Wattnode) Hot water supply temperature Hot water return temperature 5 min kwh 5 min Degrees F 5 min Degrees F x x x x Total system water flow 5 min Gallons
Electric Water Heater (Baseline Domes) Total unit current 5 min Amps x x 4. DATA ANALYSIS AND RESULTS The goals of this data analysis were to determine the impact of the retrofit technologies on the energy performance of the Domes, normalizing for any extraneous circumstances between the baseline and retrofitted residences, and to evaluate the performance of the installed heating/cooling and water heating systems. 4.1 HEATING PERFORMANCE EQUATIONS A critical aspect of the energy performance analysis for this project is the heating performance of the heat pumps in the retrofitted Domes. Heat pump heating performance is typically quantified using the Coefficient of Performance (COP). COP is defined as the dimensionless ratio of useable thermal energy to the energy used to operate the system. Specifically, COP is calculated using the following equations: EQUATION 1. COEFFICIENT OF PERFORMANCE (COP) COP = HC /(3.412 DMD) Where HC is the heating capacity of the unit (Equation 2). DMD is the electrical demand of the heat pump as measured by the Wattnode on each unit. EQUATION 2. HEATING CAPACITY (HC) HC = HE + (3.412 DMD ) HE is the heat of extraction (Equation 3). EQUATION 3. HEAT OF EXTRACTION (HE) HE = 500 GPM ΔT GPM is the ground water flow as measured by the flow meter. ΔT is the temperature difference between the ground water supply and return. For the air source heat pump in Dome 15 the calculation is the same except that heat of extraction is calculated differently: HE = 1.08 CFM ΔT CFM is the rated air flow of the unit. ΔT is the temperature difference between the return air and the supply air. 4.2 COOLING PERFORMANCE EQUATIONS Determining the cooling performance of the heat pumps in the retrofitted Domes is also a vital component of the overall energy performance analysis. Heat pump cooling performance is typically quantified using the Energy Efficiency Rating (EER). EER is defined as the ratio of the flow rate of useable thermal energy in Btu/h to the electric power used to operate the system in W. Specifically, EER is calculated using the following equations:
EQUATION 4. ENERGY EFFICIENCY RATING (EER) EER = COP 3.412 Where COP = HC /(3.412 DMD) For cooling applications HC = HE HE = 500 GPM ΔT DMD is the electrical demand of the heat pump as measured by the Wattnode on each unit. GPM is the ground water flow as measured by the flow meter. ΔT is the temperature difference between the ground water supply and return. For the packaged terminal heat pump in Dome 15 the calculation is the same except that: HE = 1.08 CFM ΔT CFM is the rated air flow of the unit. ΔT is the temperature difference between the return air and the supply air. It should be noted that the cooling and heating performance equations are only meaningful when the heat pumps are actively cooling and/or heating. In other words, the compressor and fan should be running. During this analysis, the raw data are filtered for times when the compressor is actually running before any efficiency calculations are made. 4.3 HEATING PERFORMANCE RESULTS AND DISCUSSION Because of data quality issues during the first quarter of the study, heating performance data is not available for the beginning of 2014. Specifically, there were issues with the location of the ground loop temperature sensors in both Domes 13 and 14 as well as the supply air temperature sensor in Dome 15. The sensors appeared to be picking up heat from the compressor motors and skewing the data. This issue was identified and corrected on May 22, 2014. Instead, data from November 1 st 2014 and February 28 th 2015 was analyzed. Using the COP equations presented in section 4.1, the average COP (when the unit was actively heating) between November 1 st 2014 and February 28 th 2015 was calculated with the following results: Dome 13 Ground Source HP Heating COP=3.2 Dome 14 Ground Source HP Heating COP=3.0 Dome 15 Packaged Terminal HP Heating COP=3.2 These numbers indicate decent heating efficiency although not as efficient as expected. This is potentially due to the fluctuating entering water temperatures. The ground source heat pumps were expected to operate at a rated COP of 3.8 as estimated by bench testing performed by Enertech. The COP rating is for 45 F entering water temperature, and while the average entering water temperature for the units was indeed 45 F, the minimum temperature was as low as 30 F. As an example, during a relatively long operating period in February, the entering water temperature was initially 62 F, however after two hours, it dropped to 32 F. Because the unit is so much less efficient at lower entering water temperatures, the average unit COP is reduced. The packaged terminal heat pump came from the manufacturer with a rated COP of 3.2, which the monitoring confirms as accurate in this application. The following three figures show the performance of each heat pump during a relatively cold day in the winter:
FIGURE 7. DOME 13 GSHP HEATING SYSTEM TEMPERATURES FIGURE 8. DOME 13 GSHP HEATING PERFORMANCE
Figure 7 and Figure 8 above show the heating performance of the Dome 13 ground source heat pump during a relatively cold day of the winter. Looking at the indoor air temperature (red line) reveals that the unit is effective at heating the indoor space. As soon as the unit comes on (purple line) the indoor air temperature rises quickly. Ground water delta T s are averaging around 5 F and the heating COP remains fairly constant around 3. This all indicates that the ground loop heat pump is operating properly. It is interesting to note that rather than keeping a consistent set point at the thermostat, the indoor air temperature is allowed to drop as low 60 F before the heat pump is turned on. It appears that the occupants in this dome turn the thermostat on and off depending on whether they are home.
FIGURE 9. DOME 14 GSHP HEATING SYSTEM TEMPERATURES FIGURE 10. DOME 14 GSHP HEATING PERFORMANCE
Figure 9 and Figure 10 show the performance of the Dome 14 heat pump during the same cold day in January that was shown for Dome 13. This unit appears to operate in a similar manner to the Dome 13 unit. Indoor air temperature responds quickly when the heat pump comes on and the ground water delta T averages around 5 F. Again, the residents in this Dome allow the indoor air temperature to drop below 60 F before turning on the heat pump. Interestingly, this unit generally draws around 20% more power than the Dome 13 unit. This may be due to greater head pressure in the ground loop, or there may be some difference in the way the water coils in the unit are piped. Overall, this unit appears to be working correctly and efficiently.
FIGURE 11. DOME 15 PTHP HEATING SYSTEM TEMPERATURES FIGURE 12. DOME 15 PTHP HEATING PERFORMANCE
Figure 11 and Figure 12 show the performance of the Dome 15 packaged terminal heat pump, again on the same day in January. This unit appears to be cycling more often than the other two heat pumps. This type of operation may indicate that the occupants are leaving the thermostat at one set point for longer periods rather than turning it on and off. Another interesting observation is that this unit is drawing considerably more power than the two ground loop heat pumps despite having the same rated capacity and similar efficiency ratings. It does appear that the unit is operating correctly since the indoor air temperature increases when the unit runs and the COP value is calculated to be around 3 for the entire heating season. 4.4 COOLING PERFORMANCE RESULTS AND DISCUSSION Using the EER equations, the average EER (when the unit is actively cooling) between February 7th and September 30th was calculated with the following results: Dome 13 Ground Source HP Cooling EER=11.9 Dome 14 Ground Source HP Cooling EER =12.1 Dome 15 Packaged Terminal HP Cooling EER =10.6 These numbers indicate decent cooling efficiency although potentially not as efficient as expected. For reference, the rated EER for both of the ground source heat pumps was 18.8 as estimated by bench testing performed by Enertech for a 90 F entering water temperature. In the field, the average entering water temperatures were around 110 F and reached temperatures as high as 124 F. The packaged terminal heat pump came from the manufacturer with a rated EER of 10.7. This unit likely comes closer to the rated EER because the actual conditions it experienced were, on average, closer to the rated conditions. The following figures show the performance of each heat pump during the summer:
FIGURE 13. DOME 13 GSHP COOLING SYSTEM TEMPERATURES FIGURE 14. DOME 13 GSHP COOLING PERFORMANCE
Figure 13 and Figure 14 show the performance of the Dome 13 ground source heat pump during a hot day of the summer. Looking at the indoor air temperature (red line) reveals that the unit is effective at cooling the indoor space even with outside air temperatures well over 100 F. As soon as the unit comes on (purple line) the indoor air temperature drops quickly. Ground water Delta T s are averaging around 8 F and the Cooling EER remains fairly constant. This represents a properly operating ground source heat pump. It is interesting to note that the temperature set point is relatively high, between 75-80 F, resulting in reduced overall cooling energy and higher unit efficiency.
FIGURE 15. DOME 14 GSHP COOLING SYSTEM TEMPERATURES FIGURE 16. DOME 14 GSHP COOLING PERFORMANCE
Figure 15 and Figure 16 show the performance of the Dome 14 heat pump during a cooling period in the summer. A different day is shown for this unit than for the previous figure because the residents do not always use their cooling on the same days, and this day was a good example of the unit running for much of the day. Compared to the heat pump in Dome 13, the unit does not appear to be affecting indoor temperature very effectively. This is evident by looking the indoor air temperature (red line); there are not significant reductions in temperature when the unit comes on (purple line). Additionally, the heat pump does not remain on for long periods. One possible explanation for this behavior is that the arrangement of the furniture in the room is such that the heat pump discharge air is being directed toward the thermostat. This would cause the short cycling while preventing the actual space temperature from being affected. That said, on days when the heat pumps are actually used, the measured space temperature is in fact cooler than the indoor temperature recorded in the baseline Domes during the same time period as demonstrated in Figure 20. During the times when the heat pumps remain off, space temperatures are similar between the retrofitted and baseline Domes.
FIGURE 17. DOME 15 PTHP COOLING SYSTEM TEMPERATURES FIGURE 18. DOME 15 PTHP COOLING PERFORMANCE
Figure 17 and Figure 18 show the performance of the Dome 15 packaged terminal heat pump, again for a different time period over the summer. The operating behavior of this unit is different from the ground loop heat pumps in that the unit remains on for one entire cooling period and does not cycle as often. This may be due to furniture or other items blocking the discharge of the unit preventing the cool air from circulating throughout the Dome. Regardless of the fact that the heat pump is not cycling, the unit appears to be working correctly and efficiently. Indoor air temperatures appear to be reduced when the unit runs and the EER value is consistently around 10. 4.5 OVERALL ENERGY USE COMPARISON The following tables provide an interesting energy use comparison spanning from March 2014 through February 2015: TABLE 3. HEATING ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 (Space Heater) 225 28 0 0 0 0 0 0 165 243 343 97 1100 Dome 8 (Space Heater) 26 5 0 0 0 0 0 0 12 118 79 0 241 Dome 10 (Space Heater) 2 43 2 1 0 0 0 0 184 96 307 202 837 Dome 13 (GSHP) 8 25 13 0 0 0 0 1 30 98 130 53 357 Dome 14 (GSHP) 103 49 7 2 1 1 5 17 114 165 233 126 824 Dome 15 (PTHP) 85 24 0 0 0 0 0 3 76 263 239 18 708 Table 3 shows month by month heating energy consumption for each Dome. A closer look at the data confirms that heating energy is only consumed during the winter months and is generally zero in the summer. This table demonstrates that heating energy consumption is very diverse. For example, Dome 6 used more than 5 times the amount of heating energy as Dome 8 even though they should have roughly equal heating efficiency. This difference is mostly dependent on occupant behavior, for instance on a relatively cool evening on 9/27/14, Dome 6 had an indoor heating temperature of 75 F while Dome 8 had an indoor temperature of only 66 F. A similar observation can be made of the three retrofitted domes, which show a large variation in heating energy use. Over the course of the heating season, there are several instances when indoor temperatures differ by up to 10 F, which is almost entirely due to different thermostat setpoints. It is worthwhile to note that on average, the three retrofitted Domes use about 15% less heating energy than the three baseline Domes despite having higher average indoor air temperatures in the winter (see Figure 21). This is likely partially thanks to the contribution from the slab insulation. TABLE 4. COOLING ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 (No Cooling) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 0 Dome 8 (No Cooling) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 0
Dome 10 (No Cooling) N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 0 Dome 13 (GSHP) 0 2 5 47 42 24 22 14 2 1 0 0 159 Dome 14 (GSHP) 1 3 27 111 55 47 50 6 0 0 0 0 300 Dome 15 (PTHP) 5 11 33 57 77 62 40 15 5 4 4 6 320 Table 4 shows month by month cooling energy consumption for each Dome. It is fairly clear that cooling energy is only consumed in the summer and is generally zero in the winter. Because the baseline Domes do not have any mechanical cooling, no meaningful comparison can be made between baseline and retrofitted Domes. However, it is evident that, like heating, the cooling energy consumption is more affected by occupant behavior (i.e. thermostat setpoint) than any other factor. As discussed in Section 4.4, Domes 13 and 14 have very similar cooling efficiencies, however the total energy use is significantly different between the two. The heat pump data show that the Dome 14 unit was running about twice as often as the Dome 13 unit during the summer of 2014. TABLE 5. PLUGS/LIGHTS ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 (Baseline) 37 26 20 28 40 30 14 50 42 47 50 44 429 Dome 8 (Partial Baseline) 32 30 25 44 60 60 48 40 43 48 44 36 509 Dome 10 (Baseline) 53 54 45 42 31 30 40 56 47 45 101 63 606 Dome 13 (Retrofitted) 27 44 45 48 39 56 30 63 57 50 86 86 632 Dome 14 (Retrofitted) 73 57 65 56 55 47 43 42 51 50 50 52 640 Dome 15 (Retrofitted) 53 44 41 64 73 67 53 43 31 42 34 28 573 Table 5 shows month by month plugs and lights energy consumption for each Dome. The retrofitted Domes have upgraded lighting which helps to reduce lighting energy use. However, because there is so little installed lighting, those savings are overshadowed by plug load energy use including plug in lighting. Yet again, plug loads are entirely dependent on occupant behavior and there is little difference between the baseline and retrofitted Domes. Energy consumption is generally equal month to month and Dome to Dome. The average annual consumption in this category is 565kWh with a standard deviation of only 80kWh. TABLE 6. RANGE ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 36 27 17 20 13 30 7 32 20 19 40 40 300 Dome 8 26 24 17 17 14 19 11 19 13 15 13 14 203 Dome 10 46 38 22 17 1 1 13 28 21 24 20 31 264 Dome 13 90 91 66 56 59 60 26 58 63 52 46 45 712 Dome 14 26 34 20 18 16 15 20 33 43 32 25 21 303 Dome 15 31 32 26 30 28 22 32 32 40 35 76 55 440
Table 6 shows month by month electric range energy consumption for each Dome. As expected, this end- use is very similar to the plugs and lights in that consumption is relatively constant month to month. Dome 13 consumed by far the most energy in this category, coming in almost two standard deviations above the average of 370 kwh/yr. TABLE 7. DHW ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 (Electric WH) 147 86 54 55 60 51 61 134 143 121 159 149 1219 Dome 8 (Heat Pump WH) 63 50 49 44 38 36 35 28 44 77 65 41 569 Dome 10 (Electric WH) 150 153 121 86 51 68 124 101 119 113 150 121 1357 Dome 13 (Shared Heat Pump) 215 296 221 207 280 248 189 241 301 387 445 390 3420 Dome 14 (Shared Heat Pump) 215 296 221 207 280 248 189 241 301 387 445 390 3420 Dome 15 (Shared Heat Pump) 215 296 221 207 280 248 189 241 301 387 445 390 3420 Table 7 shows month by month domestic hot water energy consumption for each Dome. The Domes 13-15 have a shared hot water system. Although there are separate hot water flow meters in place for each Dome, unfortunately the recirculation flow is captured as well and so the data from those meters is not representative of the actual amount of water consumed at each Dome. Rather than using flow data to disaggregate the domestic hot water energy, the individual Dome energy use is simply estimated as one third of the total energy usage for the whole system. These numbers show a significant increase in energy use for DHW in the retrofitted Domes, which is disappointing given the system design which should theoretically be much more efficient. A more in- depth discussion of this result follows in Section 4.6. Interestingly, the Dome 8 standalone hot water heat pump appears to be the best performer, using less than half of the energy use of the two other baseline Domes. It should be noted that hot water energy use is dependent on actual hot water consumption which we do not have data for in this case. Another valuable observation is that generally more hot water energy is used in the winter months. TABLE 8. TOTAL ENERGY (KWH) USE COMPARISON MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN '15 FEB '15 TOTAL Dome 6 (Baseline) 445 167 90 103 113 111 82 216 370 430 592 329 3048 Dome 10 (Baseline) 148 109 91 106 112 115 94 86 111 258 201 91 1523 Dome 8 (Partial Baseline) 252 288 190 147 83 99 178 184 371 277 579 416 3064 Dome 13 340 458 349 359 419 387 267 379 453 588 707 575 5281
(Retrofitted) Dome 14 (Retrofitted) 419 439 340 394 407 357 308 339 510 633 754 589 5488 Dome 15 (Retrofitted) 390 405 321 358 458 399 313 335 454 731 799 498 5461 Table 8 is a total of the monthly energy use in each Dome. The most surprising result this table reveals is that on average, the retrofitted Domes used significantly more energy than the baseline Domes. This is somewhat alarming; however it can be explained by a careful examination of the monitored performance data. FIGURE 19. ACTUAL ENERGY BY END-USE (MARCH 2014 FEBRUARY 2015) Figure 19 summarizes the consumption tables. This helps to show that the largest single contributor to energy use is domestic hot water heating. Even in the baseline Domes, this is the largest end- use, however the retrofitted Domes appear to have a fairly significant issue with more than half of all the energy consumed going to domestic hot water heating. The Dome 8 heat pump water heater shows itself to be the best performer by a large margin. Further discussion of the hot water heating issue follows in section 4.6. Another contributor to the retrofitted Dome energy use is cooling energy. Because the baseline Domes do not have any mechanical cooling, it is expected that the retrofitted Domes would use more HVAC energy during the summer. However, as Figure 20 illustrates, the retrofitted Domes do generally stay cooler inside and peak temperatures were lower.
FIGURE 20. AVERAGE I NDOOR TEMPERATURES Figure 20 shows the average indoor air temperature in the Domes during a few days of the summer. This time period was selected to demonstrate that the cooling equipment in the retrofitted Domes is working effectively to reduce indoor air temperatures. During this period the baseline Domes reach indoor temperatures up to 93 F which is of course outside generally accepted comfort ranges. On average, the retrofitted Domes are 5-10 F cooler during the peak of the day. It should be noted that the Domes experience limited summertime occupancy, and so there are many days when there is not temperature difference between the baseline and retrofitted Domes because they are unoccupied and the heat pumps are off. While mechanical cooling does increase energy use, it is justifiable as it provides a more comfortable and reasonable living space.
FIGURE 21. DOME INDOOR TEMPERATURE COMPARISON Figure 21 shows the indoor temperatures of the Domes plotted against outside air temperature for the entire year. This illustrates the fact that the retrofitted Domes are maintaining more comfortable indoor air temperatures in both the heating and cooling seasons. This is especially encouraging on the heating side since the retrofitted Domes used less heating energy overall than the baseline Domes. 4.6 DOMESTIC HOT WATER ANALYSIS It is disappointing to learn that the domestic hot water energy usage in the retrofitted Domes is significantly higher than in the baseline Domes. This is unexpected, because the baseline Domes have basic electric resistance water heaters compared with the shared DHW system which is a heat pump with a rated COP of up to 3.0 (3 times more efficient than the baseline.) However, as the following discussion concludes, there may be a way to correct this issue and improve the energy performance of the shared DHW system.
FIGURE 22. DOMESTIC HOT WATER COMPARISON Figure 22 shows various critical attributes associated with the shared domestic hot water system serving the retrofitted Domes. This data spans two days in August 2014. The water heaters in the baseline Domes (blue line) typically cycle on and off for very short periods throughout the day, while the shared DHW system (yellow line) stays on for much longer periods and has frequent spikes on top of that. The plateaus of power consumption around 250 W indicate times when the heat pump is operating, and the spikes in power use are from the electric resistance backup supplementing the heat pump. Looking at the data between 8/21-8/22, there are multiple instances where the heat pump operates as an electric resistance heater whenever any water flows at all. This is a feature of the heat pump operation which provides a faster recovery time at the expense of efficiency. Part of the reason for the unexpectedly high energy use is that the shared system for the retrofitted Domes has a recirculation loop running underground which, despite having some insulation, is constantly losing heat to the ground. On average, the water is returning from the recirculation loop around 0.5-1 F cooler than the supply temperature even when no hot water is being used. The cooler return water temperature causes the heat pump to run in order to bring the water back to set point which results in the heat pump operating for much of the day. The other major factor is that the electric resistance backup frequently cycles on. Whenever hot water is used, cold makeup water is mixed into the return line and the water heater employs the backup heater to recover more quickly. In the third quarter of 2014, the extra energy used by the backup heating element accounted for approximately 45% of the total hot water energy use. Fortunately, there are controls provided with the DHW heat pump which may help alleviate the current high energy usage. Specifically, the operating mode can be changed to eliminate the electric resistance backup. This particular water heater has four operating modes: Efficiency, Hybrid, Electric, and Vacation mode. The
unit is currently in Hybrid mode, which has shown itself to use the backup electric heat more frequently than desirable. On October 29th, 2014 the heat pump operating mode was changed to Efficiency mode which should have eliminated the backup electric resistance heater. Given that the backup heating element accounted for 45% of the energy use between June and September, this should have helped by saving at least that much energy. However, despite this change, there was no reduction in electric resistance backup operation as shown in the following figure. FIGURE 23. DHW OPERATION OCTOBER Figure 23 is the same as Figure 22 but shows a three day period around the time when the operating mode of the shared hot water heat pump was changed to efficiency mode. Looking at the heat pump power (yellow line) there does not appear to be any improvement in the operation of the heat pump during this time period. In fact, it appears that the heat pump is now operating in an electric resistance only mode. This is evident because whenever there is any power draw (yellow line) it is at ~1400 watts which is the power draw of the electric resistance heater. The heat pump should only draw ~333 watts per Dome or less. Increased hot water consumption may explain some of this type of operation, however there is not a single time when the heat pump operates without the backup heater during this two day period. This type of operation has continued through the monitoring period, indicating that nothing has changed. The most logical explanation for this type of operation is that the heat pump compressor has tripped out or otherwise failed. It is strongly recommended that the heat pump be serviced by a qualified technician, then left in the Efficiency mode.
5. CONCLUSION Over the course of the year it has become clear that overall, the new mechanical equipment installed for the retrofitted Domes has resulted in increased energy usage. While this result is disappointing, further investigation and analysis has revealed reasonable explanations and possible solutions. The majority of this energy is attributed to the shared DHW system which may be partially addressable by repairing the heat pump section of the water heater and keeping the operating mode as Efficiency as described in the user manual. The additional cooling energy in the retrofitted Domes is also a contributing factor which cannot be compared to the baseline Domes which have no form of mechanical cooling. Of course, it should be noted that the increased cooling energy does provide increased comfort and more appropriate living conditions. Additionally, the average cooling efficiency has shown to be relatively good, and the cooling energy use is not beyond what would reasonably be expected for any comparable air conditioning system. The heating season has shown that the heat pumps are not only using less energy than the baseline electric resistance heaters, but are also keeping the Domes warmer and more comfortable to live in. Finally, it should be noted that the small sample size in this study does make it difficult to extrapolate the results to a larger scale. Things such as occupant behavior and occupancy have a large effect on the results and thus skew the data making more concrete conclusions difficult. Some of the more important findings are listed below. The heat pump water heater serving Dome 8 operates very efficiently, using less than half the energy as the other baseline Domes. The shared domestic hot water heat pump appears to have a failed compressor or some other issue causing it to operate very inefficiently (using electric resistance only). While there may still be potential for this system to run efficiently, this installation has demonstrated that special care and attention should be paid to heat pump water heaters. Occupant behavior has the greatest effect on HVAC and plug energy use and should be considered as an area of focus for further studies. The efficiency of the two ground source heat pumps does appear to be slightly improved over the packaged terminal heat pump, providing cooling around 10% more efficiently, however installation was significantly more costly and involved. The heat pumps are experimental proto- type units and may have room for improvement.
APPENDIX A. DRAWINGS AND PRODUCT INFORMATION