Evaluating Domestic Water Heater Performance for
NY Homes
Draft Final Report
Prepared for:
New York State Energy Research and Development Authority
Albany, NY
Gregory Pedrick, Joseph Borowiec
Project Managers
Prepared by:
Syracuse Center of Excellence in Environmental and Energy Systems
Syracuse, NY
and
CDH Energy Corp.
Cazenovia, NY
Hugh I. Henderson, Jr., Edward Bogucz, Jeremy Wade
NYSERDA Report xx-xx
NYSERDA Contract 15606
April 2015
Notice
This report was prepared by the Syracuse COE and CDH Energy Corp. in the course of performing work contracted
for and sponsored by the New York State Energy Research and Development Authority (hereafter “NYSERDA”).
The opinions expressed in this report do not necessarily reflect those of NYSERDA or the State of New York, and
reference to any specific product, service, process, or method does not constitute an implied or expressed
recommendation or endorsement of it. Further, NYSERDA, the State of New York, and the contractor make no
warranties or representations, expressed or implied, as to the fitness for particular purpose or merchantability of any
product, apparatus, or service, or the usefulness, completeness, or accuracy of any processes, methods, or other
information contained, described, disclosed, or referred to in this report. NYSERDA, the State of New York, and the
contractor make no representation that the use of any product, apparatus, process, method, or other information will
not infringe privately owned rights and will assume no liability for any loss, injury, or damage resulting from, or
occurring in connection with, the use of information contained, described, disclosed, or referred to in this report.
NYSERDA makes every effort to provide accurate information about copyright owners and related matters in the
reports we publish. Contractors are responsible for determining and satisfying copyright or other use restrictions
regarding the content of reports that they write, in compliance with NYSERDA’s policies and federal law. If you are
the copyright owner and believe a NYSERDA report has not properly attributed your work to you or has used it
without permission, please email print@nyserda.ny.gov.
ii
Executive Summary
This study evaluated a wide array of issues related to the performance of water heaters in New York homes. The
study looked at the performance of water heater products ranging from standard gas water heater tanks through high
efficiency gas-fired options. It also looked at solar and heat pump options as an alternative to standard electric water
heaters. We used laboratory testing to make comparisons between different systems and to discern the impact of the
new water use profiles in the forthcoming DOE Test Procedure. Because the water use pattern can have a
significant impact on performance, we also measured hot water flow in several homes to understand the magnitude
and timing of hot water use. Data were also collected to quantify hot water waste due to the distribution system in
several homes. The findings of this research effort as well as more general guidance about water heating
technologies are being communicated to a broad audience of homeowners, installers and other stakeholders via web
site at http://dhw.syracusecoe.org/.
Side-by-side laboratory testing evaluated the impact that different water use profiles have on water heater efficiency.
The forthcoming UEF descriptor from the new test procedure will provide a more accurate rating of the performance
of tankless water heaters; tankless unit performance had been slightly over predicted by the current EF test and
rating procedure. High efficiency products such as the gas-fired condensing tanks and hybrid products – i.e.,
tankless units with a small amount of storage – will now be rated under the new UEF descriptor which will provide a
much more accurate representation of actual efficiency.
A side-by-side comparison of different solar hot water systems showed they had similar performance resulting in a
solar fraction of 0.46 to 0.47 across the year – providing savings of about one half compared to a standard water
heater. This performance was in line with the SRCC-predicted solar fraction of 0.56 for Syracuse.
Field measurements of hot water flow in 18 homes determined the magnitude and timing of hot water use. The
median hot water use was 45 gal/day, which was closely aligned with other national studies of hot water use. The
median two-period household used 39 gallons per day and median 4-person household used 66 gallons per day. We
found that hot water use varies by more than a factor of two on a day-to-day basis for any given house. Also, hot
water use varies seasonally, especially for large households where showers become a large portion of the mix of
total hot water use.
Our measurement of hot water waste – that is, previously-heated water that is run down the drain waiting for an
acceptable temperature at the fixture – was about 9% to 25% in five houses. In two houses we installed recirculation
pumps with return lines to the tank inlet to reduce hot water waste. We found that, while these systems reduced hot
water waste, they also increase thermal losses from the system and therefore increased energy use.
iii
Heat pump water heaters (HPWHs) are promising technology for hundreds of thousands single family NY homes
that use electric water heating. Rural and suburban homes with basements are an especially good application. We
found that the actual annual efficiency was much lower than the value predicted by the EF (or the forthcoming UEF)
due to both relatively cold basement temperatures and occasional operation of resistance elements after a large water
draw. However, the HPWH still reduced electric consumption by a nearly factor of two compared to a standard
electric water heater. Product improvements are likely to further increase these energy savings.
iv
Acknowledgements
We wish to acknowledge and thank several staff at the Syracuse Center of Excellence (COE) for contributing to this
project. Thanks to Ana Fernandez for helping to indentify and recruit field test sites from the Near Westside
neighborhood to participate in this study as well as Chris Straile who coordinated the installation of the DHW
systems at the laboratory. Also thanks to James Alfiere and his team at the Syracuse University Construction Group
for overseeing design and construction of the laboratory infrastructure.
We also wish to thank several consultants who provided technical advice and assistance on the project including
Thomas Butcher from Brookhaven National Laboratory, Carlos Colon from the Florida Solar Energy Center, and
Pete Skinner of E2G Solar.
Finally, thanks to the homeowners who agreed to participate in this field test study.
v
Table of Contents
Notice ......................................................................................................................................... ii
Executive Summary ....................................................................................................................iii
Acknowledgements ..................................................................................................................... v
Table of Contents........................................................................................................................vi
List of Figures.............................................................................................................................ix
List of Tables ..............................................................................................................................xi
Acronyms and Abbreviations .....................................................................................................xii
1
2
Introduction ........................................................................................................................ 1
1.1
Background ..................................................................................................................................... 1
1.2
Goals .............................................................................................................................................. 1
1.3
Approach ........................................................................................................................................ 1
1.3.1
Technology Evaluation ............................................................................................................. 2
1.3.2
Laboratory Test ....................................................................................................................... 2
1.3.3
Field Testing ........................................................................................................................... 2
1.3.4
Communicating Results ............................................................................................................ 2
Water Heater Technologies .................................................................................................. 3
Descriptions of Water Heater Technologies ....................................................................................... 3
2.1
2.1.1
Standard Gas Storage ............................................................................................................... 4
2.1.2
High Efficiency (HE) Gas Storage: Non-Condensing .................................................................. 4
2.1.3
High Efficiency (HE) Gas Storage: Condensing ......................................................................... 5
2.1.4
Tankless Gas Water Heater ....................................................................................................... 5
2.1.5
Hybrid Gas .............................................................................................................................. 6
2.1.6
Standard Electric Storage ......................................................................................................... 6
2.1.7
Heat Pump Water Heaters ......................................................................................................... 7
2.1.8
Electric Tankless Water Heater ................................................................................................. 8
Other DHW Approaches .................................................................................................................. 8
2.2
2.2.1
Indirect Tank Systems .............................................................................................................. 8
2.2.2
Ground Source Heat Pump Systems .......................................................................................... 8
2.2.3
Solar Water Heaters ................................................................................................................. 9
Collector Type ..................................................................................................................................... 9
Circulation Approach ........................................................................................................................... 9
2.3
3
Systems Selected for Testing .......................................................................................................... 10
Test Approach and Monitoring Details ................................................................................12
vi
Laboratory Test Approach and Setup ............................................................................................... 12
3.1
3.1.1
Laboratory Setup ................................................................................................................... 14
3.1.2
Instrumentation and Measurements ......................................................................................... 14
3.1.3
Data Analysis ........................................................................................................................ 16
Field Testing Approach .................................................................................................................. 17
3.2
3.2.1
Instrumentation and Measurements at Field Test Sites .............................................................. 18
Water-Use Only Sites ......................................................................................................................... 18
Detailed / BA Sites ............................................................................................................................ 19
3.2.2
4
Laboratory Test Results ......................................................................................................23
4.1
Comparing Measured Laboratory Performance to Energy Factor ....................................................... 23
4.2
Impact of NEW Proposed Hot Water Use Profiles ............................................................................ 27
4.2.1
Proposed Normal Use Draw Profile ......................................................................................... 28
4.2.2
Proposed Low Use Draw Profile ............................................................................................. 31
4.2.3
Proposed High Use Draw Profile ............................................................................................. 34
4.3
Comparing Performance with Different Usage Profiles ..................................................................... 39
4.4
Standby Losses and Parasitic Power ................................................................................................ 44
4.4.1
Standby Losses ...................................................................................................................... 44
4.4.2
Parasitic Electric Loads .......................................................................................................... 47
4.5
5
Data Analysis Procedures ....................................................................................................... 21
Comparing Performance of Solar Systems ....................................................................................... 48
4.5.1
Initial Performance Observations (September and October 2013) ............................................... 49
4.5.2
The Wagner Drain-Back System is Fixed ................................................................................. 54
Field Test Results................................................................................................................59
5.1
Characteristics of Field Test Sites .................................................................................................... 59
5.2
Hot Water Usage ........................................................................................................................... 62
5.2.1
Hot Water Use Profiles ........................................................................................................... 62
5.2.2
Average Daily Hot Water Use ................................................................................................. 66
5.3
Entering Cold Water Temperatures ................................................................................................. 74
5.4
Hot Water Draw Events.................................................................................................................. 76
5.5
Identifying Draws and Estimating Hot Water Waste ......................................................................... 79
5.6
Field Test Site 1 – HPWH Performance ........................................................................................... 82
5.6.1
Unexpected Aspects of HPWH Performance ............................................................................ 90
5.6.2
HPWH Interactions with the Space .......................................................................................... 92
5.6.3
Impact of Recirculation Pump on System Performance ............................................................. 99
5.7
Field Test Site 2 – Condensing Storage Tank ................................................................................. 100
vii
Field Test Site 3 – Gas-Fired Condensing Tankless......................................................................... 104
5.8
6
Conclusions and Lessons ...................................................................................................106
Summary of Findings and Conclusions .......................................................................................... 106
6.1
6.1.1
Impact of Draw Profile on Efficiency .................................................................................... 106
6.1.2
Side-by-Side Solar Testing ................................................................................................... 107
6.1.3
Hot Water Use Patterns in Test Homes .................................................................................. 108
6.1.4
Disaggregating Hot Water Use and Estimating Hot Water Waste ............................................. 109
6.1.5
HPWH Field Performance .................................................................................................... 109
6.1.6
Impact of Switching to Tankless Unit .................................................................................... 110
6.1.7
Field Performance of High Efficiency Condensing Tank ......................................................... 110
6.1.8
Recirculation Pumps to Reduce Hot Water Waste ................................................................... 110
Lessons Learned .......................................................................................................................... 110
6.2
7
6.2.1
One Water Use Profile May Not Fully Characterize Performance ............................................ 110
6.2.2
Controls and Tank Size Matter for HPWH ............................................................................. 111
6.2.3
HPWH is an Energy Efficient Option in Rural Homes ............................................................ 111
References ........................................................................................................................112
Appendix A


Summary of Field Site Characteristics, Survey Responses, and Monitoring and Data Collection Details for
18 Filed Test Sites
Sites Descriptions for Five Detailed Field Test Sites
Appendix B


Histograms of Water Use and Events per Day at each Site
Histograms of Event Volume and Duration at each Site
Appendix C

Daily Hot Water Use Profiles for Each Site Based on 15-minute Data
viii
List of Figures
Figure 1. Annual Trend of Cold Water Inlet Temperature in the City of Syracuse................................................ 13
Figure 2. Schematic Showing Sensor Locations ................................................................................................ 15
Figure 3. Schematic Showing Solar System on 4th Floor and Roof .................................................................... 15
Figure 4. Photo of Laboratory Test Setup on Fourth Floor of Syracuse COE ....................................................... 16
Figure 5. Schematic Showing Water Heater System and Distribution Piping ....................................................... 20
Figure 6. Profile for a Typical Water Draw Event (from Klein 2011) ................................................................. 22
Figure 7. Water Draw Profile Imposed on Water Heaters – CURRENT DOE Draw Profile .................................. 23
Figure 8. Comparing Measured Conversion Efficiency and Energy Factors – CURRENT Draw Profile ............... 25
Figure 9. Daily Hot Water Use and Supply Temperatures – CURRENT Draw Profile ........................................ 26
Figure 10. Useful Energy vs. Inlet Water Temperature – CURRENT Draw Profile ............................................. 26
Figure 11. Water Draw Profile Imposed on Water Heaters – Normal Use Profile ................................................ 28
Figure 12. Comparing Measured Conversion Efficiency and Expected Energy Factors – Normal Use Profile ....... 30
Figure 13. Daily Hot Water Use and Supply Temperatures – Normal Use Profile ............................................... 31
Figure 14. Water Draw Profile Imposed on Water Heaters – Low Use Profile ..................................................... 32
Figure 15. Comparison of Measured Conversion Efficiency and Expected Energy Factors – Low Use Draw Profile
....................................................................................................................................................... 33
Figure 16. Daily Hot Water Use and Supply Temperatures – Low Use Profile ................................................... 34
Figure 17. Water Draw Profile Imposed on Water Heaters – High Use Profile .................................................... 35
Figure 18. Comparison of Measured Conversion Efficiency and Expected Energy Factors – High Use Profile ...... 36
Figure 19. Daily Hot Water Use and Supply Temperatures – High Use Profile ................................................... 37
Figure 20. Impact of High Use Profile on HPWH Performance ......................................................................... 38
Figure 21. Impact of Draw Profile on Conversion Efficiency – Gas-STD........................................................... 40
Figure 22. Impact of Draw Profile on Conversion Efficiency – HE-PVNT (other) .............................................. 40
Figure 23. Impact of Draw Profile on Conversion Efficiency – SOLAR ............................................................ 41
Figure 24. Impact of Draw Profile on Conversion Efficiency – SOLAR-DRAIN ................................................ 41
Figure 25. Impact of Draw Profile on Conversion Efficiency – TANKLESS ...................................................... 42
Figure 26. Impact of Draw Profile on Conversion Efficiency – HE-Cond .......................................................... 42
Figure 27. Impact of Draw Profile on Conversion Efficiency – HYBRID .......................................................... 43
Figure 28. Impact of Draw Profile on Conversion Efficiency – HPWH.............................................................. 43
Figure 29. Daily Standby Gas Use for DHW1-GAS-STD ................................................................................. 44
Figure 30. Daily Standby Gas Use for DHW2-HE-PVNT ................................................................................. 45
Figure 31. Daily Standby Gas Use for DHW6_HE-Cond .................................................................................. 45
Figure 32. Daily Standby Gas Use for DHW7-HYBRID .................................................................................. 46
Figure 33. Daily Standby Electric Use for DHW8 ............................................................................................ 46
Figure 34. Trends of Electric Consumption versus Natural Gas Input ................................................................ 47
Figure 35. Solar Fraction vs. Solar Flux for Fall 2013 ....................................................................................... 49
Figure 36. Daily Electric Power vs. Solar Flux for Fall 2013 ............................................................................. 50
Figure 37. Direct Comparison of Rheem vs. Wagner Performance for Fall 2013 ................................................. 50
Figure 38. Performance Comparison for a Moderately Sunny Day (October 1, 2013) .......................................... 52
Figure 39. Performance Comparison for a Sunny Day (September 25, 2013) ...................................................... 53
Figure 40. Performance Comparison for a Sunny Day (May 20, 2014) ............................................................... 55
Figure 41. Solar Fraction vs. Solar Flux (before and after Wagner Fix) .............................................................. 56
Figure 42. Daily Electric Power vs. Solar Flux (before and after Wagner Fix) ..................................................... 56
Figure 43. Direct Comparison of Rheem vs. Wagner Performance (before and after Wagner Fix) ......................... 57
Figure 44. Duration of Field Monitoring Period at Each Test Site ...................................................................... 59
Figure 45. Average Monthly Occupancy at Each Test Site ................................................................................ 60
ix
Figure 46. Hot Water Use Profile for 2-Person Home during a Cold Month (February 2013 & 2014) .................... 62
Figure 47. Hot Water Use Profile for 2-Person Home during a Swing Month (May 2013 & 2014) ........................ 63
Figure 48. Hot Water Use Profile for 2-Person Home during a Hot Month (August 2013 & 2014) ........................ 63
Figure 49. Hot Water Use Profile for 4-Person Home during a Cold Month (February 2013 & 2014) .................... 64
Figure 50. Hot Water Use Profile for 4-Person Home during a Swing Month (May 2013 & 2014) ........................ 65
Figure 51. Hot Water Use Profile for 4-Person Home during a Hot Month (August 2013 & 2014) ........................ 65
Figure 52. Median Hot Water Use at Each Site with a Range of Plus or Minus One Standard Deviation about the
Mean ............................................................................................................................................... 68
Figure 53. Average Daily Hot Water Use for 2-Person and 4-Person Homes ....................................................... 69
Figure 54. Average Daily Hot Water Use for 2-Person Households Compared to the Average .............................. 70
Figure 55. Average Daily Hot Water Use for 2-Person Households .................................................................... 71
Figure 56. Average Daily Hot Water Use for 2-Person Households with Each Site Shown ................................... 71
Figure 57. Average Daily Hot Water Use for 4-Person Households Compared to the Average .............................. 72
Figure 58. Average Daily Hot Water Use for 4-Person Households .................................................................... 73
Figure 59. Average Daily Hot Water Use for 4-Person Household with Each Site Shown .................................... 73
Figure 60. Daily Minimum Entering Water Temperature vs. Daily Flow-Weighted Average EWT ....................... 74
Figure 61. Seasonal Variation in Entering Water Temperature ........................................................................... 75
Figure 62. Daily Hot Water Use vs. Entering Water Temperature ...................................................................... 76
Figure 63. Hot Water Use vs. Number of Draw Events (all days through December 2014) ................................... 78
Figure 64. Hot Water Use vs. Number of Draw Events (Median for Sites through December 2014) ...................... 78
Figure 65. A Sample Hot Water Draw is Identified and Evaluated at Site 1 ......................................................... 80
Figure 66. Conversion Efficiency (or COP) as a Function of Hot Water Use ....................................................... 83
Figure 67. HPWH Input Power as a Function of Hot Water Use ........................................................................ 83
Figure 68. Plot of HPWH Performance for a Day with Element Operation - April 10, 2013 ................................. 85
Figure 69. Plot of HPWH Performance for a Day with Element Operation - April 23, 2013 ................................. 86
Figure 70. Plot of HPWH Performance for a Day with Significant Element Operation - May 21, 2013.................. 87
Figure 71. Plot of HPWH Performance for a Day with Small Amounts of Element Operation - November 12, 2013
....................................................................................................................................................... 88
Figure 72. Plot of Element Runtime as Function of Hot Water Draw Size ........................................................... 89
Figure 73. Shade Plot Showing Unexplained Periods When Lower Electric Element Runs a Small Amount .......... 90
Figure 74. Last Few Days of HPWH Compressor Operation .............................................................................. 92
Figure 75. Psychrometric Conditions in Basement – with and without HPWH Operation ..................................... 93
Figure 76. Wet Bulb and Dry Bulb in Basement – with and without HPWH Operation ........................................ 93
Figure 77. Basement Temperature and Humidity Shown with DH Operation – Entire Test Period ........................ 94
Figure 78. Basement Temperature and Humidity Shown with DH Operation – June 20, 2013 .............................. 95
Figure 79. Comparison of Daily Average Basement and Outdoor Humidity Ratios .............................................. 96
Figure 80. Effect of Outdoor Humidity Ratio on Dehumidifier Runtime ............................................................. 97
Figure 81. Shade Plot Showing Runtime of Recirculation Pump ........................................................................ 99
Figure 82. Shade Plot Showing When Recirculation Pump Operated ................................................................ 101
Figure 83. Plots of Daily Runtime for Recirculation Pump .............................................................................. 101
Figure 84. Impact of Recirculation Pump Operation on Water Heater Gas Use (when cold water is between 51 and
56°F) ............................................................................................................................................. 102
Figure 85. Variation of Efficiency with Hot Water Use for Period With Recirculation Pump Off ........................ 103
Figure 86. Comparing HW Use and Events for Conventional Tankless Systems ................................................ 104
Figure 87. Conversion Efficiency Versus Hot Water Use ................................................................................ 105
x
List of Tables
Table 1. Breakdown of Fuels Used for Water Heating in NY Homes .................................................................... 3
Table 2. High Efficiency Non-Condensing Gas Storage Units ............................................................................. 5
Table 3. High Efficiency Condensing Gas Storage Units ..................................................................................... 5
Table 4. Tankless Condensing Water Heaters ..................................................................................................... 6
Table 5. Hybrid (Small Storage) Water Heaters .................................................................................................. 6
Table 6. Heat Pump Water Heaters .................................................................................................................... 7
Table 7. DHW Systems Recommended for Testing .......................................................................................... 10
Table 8. List of Monitored Points and Instrumentation for Laboratory Testing .................................................... 16
Table 9. Summary of Testing at the Field Test Sites.......................................................................................... 18
Table 10. List of Monitored Points and Instrumentation .................................................................................... 21
Table 11. Summary of Water Heater Performance – CURRENT DOE Draw Profile ........................................... 24
Table 12. Summary New Hot Water Use Profiles for Proposed ASHRAE 118.2 Test Procedure.......................... 27
Table 13. Summary of Water Heater Performance – Normal Use Profile ........................................................... 29
Table 14. Summary of Water Heater Performance – Low Use Profile ................................................................ 32
Table 15. Summary of Water Heater Performance – High Use Profile ............................................................... 35
Table 16. Summary of Conversion Efficiencies with Different Draw Patterns .................................................... 39
Table 17. Summary of Water Heater Standby Losses ....................................................................................... 44
Table 18. Electric Use of Gas-Fired Systems .................................................................................................... 47
Table 19. Solar Systems Installed in the Laboratory .......................................................................................... 48
Table 20. Summary Impact of Fixing Wagner System....................................................................................... 57
Table 21. Monthly and Annual Performance Summary for Solar Systems ........................................................... 58
Table 22. Summary of 18 Field Test Sites ........................................................................................................ 61
Table 23. Beginning and Ending Dates for Data Collection at Sites with 15-minute Monitoring ........................... 66
Table 24. Impact of Including Vacation and Missing Days in Hot Water Use Calculation .................................... 67
Table 25. Beginning and End Dates for Event Data at Each Site with 5-second Monitoring.................................. 77
Table 26. Summary of Hot Water Draw Identification at Site 1 .......................................................................... 81
Table 27. Summary of Monthly HPWH Performance ....................................................................................... 84
Table 28. Summary of Days with Unexpected Element Operation...................................................................... 91
Table 29. Multi-linear Regression Results ........................................................................................................ 98
Table 30. Monthly Summary of Dehumidifier and HPWH Operation ................................................................. 98
Table 31. Impact of Recirculation Pump Operation on Standby Losses ............................................................. 103
xi
Acronyms and Abbreviations
ACEEE
AHRI
ASHRAE
BA
Btu
COE
COP
DOE
DH
DHW
EIA
EPA
EF
°F
FSEC
gal
gpm
GSHP
HE
HPWH
HW
HX
kW
kWh
MBtu
NEEA
NY/NYS
NYSERDA
PEX
POU
RH
SEF
SF
SRCC
TC
TE
UEF
UL
WH
American Council for an Energy Efficient Economy
Air Conditioning, Heating and Refrigeration Institute
American Society of Heating Refrigerating and Air Conditioning Engineers
Building America. Research program at DOE related to residential energy efficiency
British Thermal Units
Center of Excellence
Coefficient of Performance. Same as conversion efficiency
US Department of Energy
Dehumidifier
Domestic Hot Water
Energy Information Agency (division of US DOE)
Environmental Protection Agency
Energy Factor (DOE rating for water heaters)
Degrees Fahrenheit
Florida Solar Energy Center
gallons
gallons per minute
Ground Source Heat Pump
High Efficiency
Heat Pump Water Heater
Hot Water
Heat Exchanger
kilowatt
kilowatt-hours
thousands of Btu
Northwest Energy Efficiency Alliance
New York / New York State
New York State Energy Research and Development Authority
Polyethylene Pipe
Point of Use
Relative Humidity
Solar Energy Factor (SRCC rating for solar systems under OG-300)
Solar Fraction. Portion of load met with solar energy
Solar Rating and Certification Corporation
Thermocouple
Thermal Efficiency (efficiency rating for commercial water heater products)
Uniform Energy Factor (new DOE rating for water heaters)
Underwriters Laboratory
Water Heater
xii
1
Introduction
1.1
Background
Water heating is the second largest energy expenditure in existing homes in the Northeast, after space heating (EIA,
2013). Historically, efforts to improve residential efficiency have focused on space conditioning, often neglecting
water heating improvements. Several new or recently-refined water heating technologies are now available on the
market including: solar water heaters; gas-fired, tankless units; and heat pump water heaters. Tax credits are
currently available for some of these systems that boost consumer interest.
Current DOE rating procedures to determine the Energy Factor (EF) 1 , or efficiency, of these systems may be a poor
indicator of actual energy use, in part because the amount and timing of water use greatly impacts the performance
and relative efficiency ranking of these systems. Robust measured data is needed to help consumers, manufacturers,
and installers understand the efficiency, costs, and other impacts of both new and conventional domestic hot water
systems.
1.2
Goals
This research and demonstration project sought to answer the following questions:





1.3
How will the expected changes to the DOE Test Procedure affect published efficiency rating of water
heaters? Will it provide a more realistic representation of actual performance in New York Homes?
What is the magnitude and timing of water heating loads in NY Homes? How NY homes compare to the
national market?
What is magnitude of hot water waste in homes? Can distribution system improvements and technologies
such recirculation pumps reduce this waste?
How do the broad range the broad range of electric, gas-fired and solar water heating options compare in
NY homes? What new technologies have the most potential in NY homes?
How well do the newest energy-efficient water heating technologies work in actual homes?
Approach
To answer these questions we engaged in a research effort that included a survey of current technologies, a
laboratory testing effort to assess the impact of load profile on water heater efficiency and to compare solar and
conventional technologies, and a series of field tests to understand actual water heating loads in NY homes and test
the newest technologies under actual conditions. The findings of this research effort as well as more general
1
As defined in 10CFR430 Subpart B, Appendix E and in ASHRAE Standard 118.2
1
guidance about water heating technologies are also being communicated to a broad audience of homeowners,
installers and other stakeholders via web site.
1.3.1
Technology Evaluation
To determine which water heating technologies were most important to New York State, we teamed with the
American Council for an Energy Efficient Economy (ACEEE). ACEEE (2011) developed its “Emerging Hot Water
Technologies and Practices for Energy Efficiency as of 2011” report which is available at http://aceee.org/researchreport/a112 . Using this ACEEE evaluation as a starting point, we focused in on the emerging technologies and
approaches that were most appropriate for this NY-focused research effort. Section 2 of this report summarizes our
evaluation of the technologies and lists the systems identified for laboratory testing.
1.3.2 Laboratory Test
A side-by-side test laboratory was constructed at the Syracuse COE to test various standard and high efficiency
water heating systems appropriate for northern climates such as New York. The laboratory setup allows the
different units to be subjected to the same loads and environmental conditions to directly compare measured
performance. This setup is similar to laboratory at the Florida Solar Energy Center (FSEC) that has focused on
water heating technologies appropriate for that climate. The test setup also allowed us to study the impact of
different water use profiles on efficiency and performance. Section 3.1 provides full details of the laboratory test
setup and Section 4 presents and discusses the results.
1.3.3 Field Testing
Field testing was conducted at 18 homes to measure the magnitude and timing of hot water use in NY households.
Some sites also included enhanced data logging capabilities to capture detailed data about each hot water draw
event. Five field test sites also included temperature sensors on the distribution piping to determine the path of each
hot water draw to allow us to allocate it to each fixture or end use and ultimately determine the amount of hot water
waste. At three sites we also installed high efficiency water heaters to evaluate their performance under actual
conditions. Section 3.2 provides full details of the field test approach and Section 5 presents and discusses the field
test results.
1.3.4 Communicating Results
As a culmination of the project, we developed a web site to communicate key findings from this research as well as
to provide general guidance to NY consumers and installers. This website is available at dhw.syracusecoe.org.
2
2
Water Heater Technologies
Sixty one percent (61%) of households in the Northeast use natural gas fueled water heaters (EIA, 2013b). A more
specific breakdown of water heating fuels for New York State households is given in the table below from the
NYSERDA Patterns and Trends – New York State Energy Profiles: 1998-2012 document based on the 2009 RECS
data (NYSERDA, 2014). The estimated average size of the heating load associated with each fuel is also given.
Electric water heating accounts for 17% of households and is typically used in rural and suburban homes as well as
apartments where loads are smaller (36 gallons per day). Natural gas is by far the most popular fuel, accounting for
61% of households. The average load size (56 gallons per day) is more in line with expectations for a single family
home. Nearly every home where natural gas is available (and used for space heating) also uses this fuel for water
heating. Fuel oil is used in 2.1 million homes for space heating, however only a portion of these homes (1.3 million)
also use it for their water heating fuel, Fuel oil is less prevalent for water heating due to the high cost of these
appliances. When fuel oil is used, it is most likely with a boiler that has an additional heat exchanger or indirect
tank for domestic water heating. The remaining 700,000 homes with oil heat presumably use electricity for water
heating. Propane is not widely used (only 3% of households) for water heating, perhaps only in cases when it is also
used for cooking fuel in rural applications.
Table 1. Breakdown of Fuels Used for Water Heating in NY Homes
No of NY
Households for
Water Heating
Avg Use / Cost
Implied1 Hot Water
Use (gal/day)
No of NY
Households for
Space Heating
Electricity
1,200,000 (17%)
2,333 kWh / $398
36
500,000 (7%)
Natural Gas
4,400,000 (61%)
200 therms / $299
56
4,100,000 (57%)
Fuel Oil
1,300,000 (18%)
120 gal / $305
55
2,100,000 (29%)
Propane
200,000 (3%)
175 gal / $545
45
200,000 (7%)
TOTAL
7,200,000
7,200,000
Notes: 1 – implied water use determined using fuel energy content, expected efficiency for each fuel and 70°F temp
rise.
2.1
Descriptions of Water Heater Technologies
Residential water heaters are classified by their input fuel and by the amount of storage capacity. The two primary
fuels used for the US market are electric and natural gas. Most natural gas units can also be converted to propane.
Oil-fired units are also available but are much less common (in New York State homes with indirect heat exchangers
are thought to be more common than direct-fired tanks).
The DOE rating procedure for Energy Factor (EF) applies to storage water heaters both gas and electric, as well as
tankless units and heat pump water heaters. To be classified as tankless the burner input must be at least 50 MBtu/h
and the storage capacity must be less than 2 gallons. EF is a measure of a water heater’s overall daily efficiency
3
based on the amount of hot water produced per unit of fuel consumed. The draw profile for the EF rating consists of
six (6) consecutive draws of 10.7 gallons at the beginning of each hour, for a total of 64.3 gallons a day.
2.1.1 Standard Gas Storage
Standard gas storage water heaters have an EF range of 0.56 to 0.62, an input rating of 30 to 75 MBtu/h, and a first
hour rating between 48 and 114 gallons per hour (AHRI, 2011). A standard gas storage water heater burns fuel in
the bottom of the tank and exhaust gases pass through a flue passage in the center of the storage tank. The water in
the tank extracts the heat from the combustion gases in the combustion chamber and flue. Standard units have a
standing pilot flame and controls that typically do not require electric power to operate.
The installed cost of a 40 gallon standard gas storage water heater with an EF below 0.60 is typically in the range of
$350 to $600 according to ACEEE (2011).
2.1.2 High Efficiency (HE) Gas Storage: Non-Condensing
There are two types of high efficiency (HE) gas storage water heaters: non-condensing and condensing. According
to the DOE (2010) a non-condensing HE gas storage unit can save as much as $246 a year with a simple payback in
as little as 10 years. As of September 1, 2010 there were 60 models available that qualified for Energy Star
certification (DOE, 2010).
HE non-condensing gas storage water heaters are similar to standard gas storage units. These HE units meet the
Energy Star criteria with an EF rating of 0.67 or greater, which is typically 12-16% more efficient than a standard
gas-fired unit. The increased efficiency is typically obtained through:






Electronic ignition: eliminating the continuous pilot light,
Improved flue baffling: increasing the surface area and turbulence to improve heat exchange,
Flue dampers: dampers close while the water heater is in standby mode; restricting off-cycle airflow that
loses heat up the flue.
Additional insulation: increasing the tank R-value to reduce heat losses to the surrounding space.
Heat traps: one-way valves installed on the hot/cold outlet/inlet pipes to prevent convective heat and mass
flow.
Forced air intake systems and power venting: improves combustion efficiencies and heat exchange.
The approximate installed cost of a HE non-condensing gas storage water heater is $1,300 (ACEEE, 2011). The
four major manufacturers of HE non-condensing gas storage water heaters listed in the table below.
4
Table 2. High Efficiency Non-Condensing Gas Storage Units
Manufacturer / Brand
Storage Size (gal)
Rating
A.O. Smith Water Products
40, 50, and 75
EF = 0.67 to 0.70
American Water Heater Company
40 or 50
EF = 0.67 to 0.70
Bradford White Corporation
40 to 65
EF = 0.67 to 0.70
Rheem Manufacturing
29 to 50
EF = 0.67 to 0.70
2.1.3 High Efficiency (HE) Gas Storage: Condensing
Condensing gas storage water heaters incorporate technology that captures latent heat of water vapor released during
combustion. A coil-like flue that spirals up through the tank increases surface area for better heat exchange.
According to ACEEE (2011) this helps to improve the efficiency a further 10% over the non-condensing units. A
list of Energy Star qualified condensing gas storage units (with EF ratings) is not yet available. These units are
generally considered by DOE to be commercial water heaters; however, smaller “commercial” condensing water
heaters are available for residential applications. The approximate installed cost of a condensing gas storage water
heater is $2,500 (ACEEE, 2011). There are currently four companies that offer condensing gas storage water
heaters that are listed in the table below.
Table 3. High Efficiency Condensing Gas Storage Units
Manufacturer / Brand
Storage Size
Rating
A.O. Smith (Vertex™)
50 gallons
90-96% thermal efficiency
American Water Heater (Polaris®)
34 to 50 gallons
95+% thermal efficiency
Bradford White (EFR)
60 gallons
95% thermal efficiency
Heat Transfer Products (Phoenix®)
55 gallons
93.7-95.1% thermal efficiency
Note: No EF rating is available for these units.
2.1.4 Tankless Gas Water Heater
Tankless water heaters are currently installed in 1.4 million homes in the Northeast region, with 400,000 estimated
to be in New York (EIA, 2013c). According to Energy Star yearly savings of up to $201 are possible with a simple
payback of 7 to 19 years (DOE, 2010). Tankless gas water heaters have a large burner (and no storage) so that water
is heated on demand. To be considered tankless by the Department of Energy the unit must have a storage capacity
no greater than 2 gallons.
Condensing Tankless Water Heater. Condensing tankless water heaters add additional heat exchanger surface to
extract more heat from the combustion gases. These condensing units have EF ratings to 0.91, compared to 0.82 for
non-condensing. The condensing tankless water heaters are approximately 37% more efficient than standard gas
storage units. The approximate installed cost of a condensing tankless unit is $2,900 plus $85 for recommended
5
annual tankless maintenance (ACEEE, 2011). Energy Star and AHRI directory list five manufacturers of
condensing gas tankless water heaters.
Table 4. Tankless Condensing Water Heaters
Manufacturer / Brand
Input
Rating
A.O. Smith Water Products
180 MBtuh to 199 MBtuh
EF = 0.91
Bosch Water Heating
175 MBtuh to 225 MBtuh
EF = 0.92 to 0.98
Navien America Inc.
150 MBtuh to 199 MBtuh
EF = 0.95 to 0.97
Noritz America Corp.
157 MBtuh to 199 MBtuh
EF = 0.91 to 0.94
Rheem Manufacturing
157 MBtuh to 199 MBtuh
EF = 0.92 to 0.94
2.1.5 Hybrid Gas
Hybrid water heaters combine the large burner of a tankless heater with a small storage tank to overcome some of
the performance short comings of tankless units. The large burner supplies a continuous supply of hot water
reducing the standby losses of a storage water heater. The small storage helps to avoid the “cold water sandwich”
problem commonly reported in tankless units and helps to eliminate the delay in hot water delivery as the system
first activates and warms (ACEEE, 2011). Condensing hybrid units have higher thermal efficiency than noncondensing hybrid heaters. The estimated installed cost of a non-condensing hybrid water heater is $1,726 with an
additional $85 a year for recommended tankless maintenance (ACEEE, 2011). The installed cost of a condensing
hybrid unit is $2,300 (ACEEE, 2011). The current manufacturers are listed below.
Table 5. Hybrid (Small Storage) Water Heaters
Manufacturer / Brand
Model
Input
EF or Efficiency
A.O. Smith Water Products
NEXT Hybrid®
100 MBtuh
90%
Grand Hall USA
Eternal Hybrid™
100 MBtuh to 195 MBtuh
EF = 0.94 to 0.96
2.1.6 Standard Electric Storage
About 24.5% of households in the Northeast and 16.7% in New York State use electric storage tanks to meet their
water heating needs (EIA, 2013c). Standard electric storage water heaters have an EF range of 0.81 to 0.95, power
input of 1.5 kW to 5.5 kW, and a first hour rating of 17 to 120 gallons per hour (AHRI, 2011). Although electric
storage water heaters are more efficient per unit of energy, the higher cost of electricity makes gas storage water
heaters cheaper to own over the lifetime of the unit (assuming gas is available). An electric storage heater has a
resistance element that directly heats the water in the tank. Electric water heaters are often used in apartments or
other confined spaces where flue gas venting is difficult or impossible.
6
2.1.7 Heat Pump Water Heaters
Heat pump water heaters (HPWH) “concentrate” heat from the ambient air to the water in the storage tank using a
vapor compression cycle. The heat pump extracts heat from the surrounding air, providing sensible cooling and
dehumidification. The AHRI directory shows that HPWHs have EFs in the range of 2.00 to 2.51, inputs of 2.5 to
5.0 kW, and a first hour ratings of 56 to 84 gallons per hour. According to the DOE (2010) a HPWH can save
homeowners around $234 a year in some climates compared to standard electric storage units, with a simple
payback of 3 years. HPWHs are available as integrated units that include storage tank as well as add-on units that
can be added to an existing tank. To have an acceptable first hour rating a resistance heating element is required.
There are currently eight manufacturers of HPWHs.
Table 6. Heat Pump Water Heaters
Manufacturer / Brand
Model
Input
Rating
A.O. Smith Water Products
Voltex®
4.5 kWh
EF = 2.33
Airgenerate LLC
AirTap™
4.0 kWh to 5.0 kWh
EF = 1.92 to 2.21
Bosch Water Heating
HP 200-1
4.5 kWh
EF = 2.20
General Electric
GeoSpring™
4.5 kWh
EF = 2.35 to 2.40
Heat Transfer Products
HTP Hybrid
4.0 kWh
EF = 2.20
Rheem Manufacturing
EcoSense
2.5 kWh to 4.0 kWh
EF = 2.00
Stiebel Eltron Inc.
Accelera® 300
2.2 kWh
EF = 2.51
USI Green Energy
Green Star
5 kWh
EF = 2.39 to 2.40
An Energy Star HPWH must have an EF of at least 2.0, a first hour rating of at least 50 gallons per hour, a warranty
of at least 6 years, and meet the required UL safety standards (Energy Star, 2011). Energy Star rates both integrated
or drop-in HPWH configurations. The expected installed price of an Energy Star HPWH is $1,600 with over $300
in base savings over a standard electric storage water heater2. There is also an additional annual savings of $20
(annual cost of operating a standalone dehumidifier) due to the dehumidification capacities of the HPWH (ACEEE,
2011).
Using a HPWH in a Northern climate often requires special considerations. Therefore, the Northwest Energy
Efficiency Alliance (NEEA, 2011) has proposed a Northern HPWH specification that includes:




2
HPWH is Energy Star qualified (EF ≥ 2)
First Hour Rating comparable to equivalent resistance water heater (60 gallons per hour)
Additional safety and service parameters for condensate management, air filters, diagnostics, and freeze
protection.
Additional exhaust ducting and noise control features for units to be installed in living spaces.
Assumed $0.1158/kWh, EF=0.90 and 4,878 kWh/yr for standard electric, EF=2.00 and 2,195 kWh/yr for HPWH
7
ACEEE estimates that the installed cost of this Northern HPWH is $1,700 (ACEEE, 2011). In our opinion some of
these features (such as the existing ducting) may not be appropriate in a New York State application where units are
likely to be mounted in a basement.
Add-on HPWH. An add-on HPWH is a standalone air source heat pump that is separately connected to a third party
tank via a pumping circuit. Cold water enters the heat pump where it is heated and pumped back to the tank. Addon heat pumps are not federally regulated. The Department of Energy does not consider them to be complete water
heaters, because they rely on third party tanks with unknown thermal properties. Therefore, a recognized rating
method does not yet exist, and have been excluded from the Energy Star Water Heater program. These units have
an estimated installed price of $800 with an additional annual savings of $20 for no longer needing a standalone
dehumidifier (ACEEE, 2011).
2.1.8 Electric Tankless Water Heater
Electric tankless water heaters are smaller units that can be located at or near the end use location or point of use
(POU), such as the sink or shower/tub. Electric POU water heaters can also be used to boost the hot water
temperatures from a central storage system or solar storage. The increased efficiencies attributed to these units are
primarily from the decrease in distribution losses of longer runs from a centralized heating system. Distribution
losses are primarily a concern in commercial application but can also be an issue in larger residential applications.
Standby losses from the tank are also eliminated. Eemax is the only manufacturer listed in the AHRI directory and a
web search found Bosch Water Heating (Powerstar and Ariston), and Stiebel Eltron Inc. Rheem produces
commercial models possibly suitable for residential applications. ACEEE estimated the installed cost of an electric
tankless unit in a remote, but extensively used bathroom to be $1,700 (ACEEE, 2011).
2.2
Other DHW Approaches
2.2.1 Indirect Tank Systems
In New York State 3.2 million households use natural gas or fuel oil boilers for space heating (EIA, 2013d). The
boiler can also provide water heating via an internal heat exchanger or an indirect tank. The internal heat exchanger
approach is a tankless approach where the boiler must fire on demand to provide tankless water heating. Indirect
tanks add another circuit or zone to boiler to heat water in an insulated tank. These tanks provide water heating at
the efficiency of the space heating boiler (excluding piping losses) and typically have internal controls to turn off the
boiler-side pumps or valves.
2.2.2 Ground Source Heat Pump Systems
Ground source heat pumps (GSHPs), also known as Geothermal or GeoExchange systems, provide space
conditioning (heating and cooling). Many of these systems can also provide domestic hot water. There are two
8
common configurations for GSHP-based water heating systems: one is a desuperheater heat exchanger that is
integrated into the GSHP unit, or the second is a triple-function GSHP unit that uses its full heating capacity to meet
DHW loads. The desuperheater uses the hot gases from the heat pump’s compressor to assist in heating the water.
During the cooling season the desuperheater provides “free” hot water, but during heating season water heating is
provided at the space heating COP. A shortcoming of the desuperheater concept is that water heating is only
provided as a consequence of space conditioning operation. The triple-function GSHP can use its full heating
capacity to provide water heating. The unit includes different refrigerant valving and controls to enable compressor
operation based on space heating, water heating, or cooling loads (thus the triple function). The estimated
incremental cost of a triple-function heat pump is $900 more than the desuperheater unit, with little to no change to
the installation costs; however other changes in the GSHP may increase the cost further (ACEEE, 2011).
2.2.3 Solar Water Heaters
The solar heater market was the only area that saw a decline in Energy Star qualified shipments from 2006 to 2009.
This is thought to be the result of overestimation of qualified products and the lack of engagement from some solar
manufacturers becoming active within Energy Star. Solar with electric backup has an annual savings around $259
and an approximate payback period of 10 years (DOE, 2010). Solar water heaters consist of three components;
collector, circulation, and storage.
Collector Type
There are three types of collectors used in solar water heating systems.
1.
2.
3.
Batch or Integral Collectors. A batch collector, also known as an integrated collector-storage system, typically
stores water in a tank attached to the collector. Heat transfer between the tank and collector is normally
achieved by passive heat transfer via a thermosyphon. This system is normally unsuitable for northern climates
where temperatures are frequently below freezing.
Flat Plate Collectors. Flat plate collectors consist of tubes attached to absorber plates which are all contained
within an insulated box.
Evacuated Tube Collectors. Evacuated tube collectors are glass tubes containing water or heat transfer liquid
inside a larger glass tube. The space between the tubes is at a vacuum and helps to improve efficiency.
Evacuated tube collectors have been shown to work at -40°F and individual tubes can be replaced.
Circulation Approach
Solar systems can use two different water circulation approaches:
1.
2.
Direct / Drain-back. Direct circulation approaches pump potable water directly through the collector where it is
heated and sent back to the tank. Usually the water “drains-back” to the tank when solar heat is not available
and when the risk of freezing is high. This approach is preferable for warmer climates where it rarely freezes,
since heat transfer is better without an intermediate fluid.
Closed Loop or In-direct. In a closed-loop system a non-freezing liquid flows through the collector and returns
to a heat exchanger in or near the storage tank. This approach is common for cold climates where the risk of
freezing is high.
9
A new in-direct, drain-back system is now available for northern climates from Wagner & Co (Secusol). The
Secusol system has a drain-back glycol loop in an in-direct heat exchanger inside the water heating tank. The
internal glycol heat exchanger has large enough volume to hold all the glycol needed to operate the system – and
store it when the system is off. This system provides the freeze protection of a glycol system but eliminates the
extra components and installation labor needed for closed loop, pressurized glycol system. It is also resistant to
overheating and stagnation issues that can degrade the glycol fluid on a sealed system.
2.3
Systems Selected for Testing
Based on the technology review above we recommended the systems listed in Table 7 for laboratory testing in this
project. The systems were chosen based on their applicability and potential in New York State homes. Table 7 lists
the characteristics for the recommended systems.
Table 7. DHW Systems Recommended for Testing
Location
1
Technology
Classification
Manufacturer and
Model
GAS-STD
A.O. Smith
Standard Gas Storage
GCVX-50-100
GAS-HE-PVNT
2
Non-condensing Power
Vent Gas Storage
A.O. Smith
GPVR-40
SOLAR
Rheem
3
Flat Plate, Glycol Solar
Hot Water
SOLPAK 3.2 RS12064BP
SOLAR-DRAIN
4a
Drain-Back Solar Hot
Water
Wagner & Co
SECUSOL 350-2
144014 28
ELECT-STD
A.O. Smith
Standard Electric Storage
ECT-52
4b
5
TANKLESS-COD
Rheem
Condensing Gas Tankless
RTGH-95DVLN
GAS-HE-COND
6
7
8
HE Condensing Gas
Storage
A.O. Smith Vertex™
GDHE-50
HYBRID
A.O. Smith
Hybrid HE Small Storage
NEXT HYB-90N
HPWH
GE
Heat Pump Water Heater
GEH50DEEDSC
EF or SEF
Input and Size
0.58
65 MBtu/h
50 gal
0.67
40 MBtu/h
40 gal
3.2
3.1
4.5 kW
120 gal
4.5 kW
92 gal
.91
4.5 kW
52 gal
0.94
199 MBtu/h
Not rated by EF
(96% efficiency)
100 MBtu/h
50 gal
not rated by EF (90%
efficiency)
100 MBtu/h
2.35
0.55-4.5 kW
50 gal
Notes: Location corresponds to the position in the laboratory. Location 1 and 2 on 1 st floor. Locations 3 to 8 on 4th
floor from west to east. SOLAR-DRAIN (4a) and ELECT-STD (4b) are combined into one system.
10
We focused on gas and electric appliances even though oil-fired water systems are relatively prevalent in New York
State. We believe that oil-fired systems are typically not stand-alone, direct-fired appliances but instead are more
commonly part of the boiler system used for space heating.
The majority of installed tankless units are non-condensing units. However, we recommend that a modern
condensing unit be laboratory tested to understand the characteristics and performance of these more advanced units.
While Heat Pump Water Heaters (HPWHs) are more widely used in southern climates they also have potential as an
efficiency improvement in NYS homes with electric water heaters. Therefore this unit was included in the test
program.
Solar Hot Water Systems are not being directly funded by the NYSERDA portion of the project but are being
separately funded by a separate DOE project at Syracuse University (DE-EE0002121). We selected two types of
systems that are appropriate for cold climates:


A conventional glycol-filled system with an indirect heat exchanger in the storage tank (Rheem)
An innovative glycol-fired drain-back system that is simpler to install and has better performance
(Wagner).
11
3
Test Approach and Monitoring Details
The approach for this project was to test water heaters in the laboratory as well as in homes under actual operating
conditions (i.e., field testing). The side-by-side laboratory testing is useful for making careful comparisons between
different systems operating under the same operating conditions. The field testing is useful for understanding and
characterizing the loads imposed on these appliances.
3.1
Laboratory Test Approach and Setup
The laboratory test setup was designed provide head-to-head comparisons of different water heating (WH) systems
under the same load conditions (i.e., hot water use patterns). The laboratory was built to include different test
stations for eight different water heaters. The side-by-side setup allows weather-dependent systems – such as solar
water heating systems and heat pump water heaters – to be directly compared to other systems on a daily basis under
different load and weather conditions. This test setup is similar to a laboratory at Florida Solar Energy Center in
Cocoa, Florida (http://infomonitors.com/hws/ and http://blog.floridaenergycenter.org/echronicle/tag/hot-watersystems-lab/). The FSEC laboratory is directly comparing WH system performance for southern climates. Our goal
was to provide a similar side-by-side assessment of actual WH system performance under conditions that are
representative of northeastern residential applications.
The laboratory setup allows for any hot water use pattern to be simulated by each of the eight test stations. A
datalogger/control system was programmed to initiate a water draw of any duration starting at any time (to the
nearest minute) in a 24-hour period. The system allows for up to 100 draw events per day. The draw rate can be
specified at 3 possible flow rates (e.g. 1, 2, or 3 gpm). This setup allows for:


a standard simulated use test according to the DOE Test Procedure to determine Energy Factor (EF), i.e.,
six draws, one hour apart, each draw at 3 gpm for 214 seconds, or
other arbitrary draw patterns to represent more typical use.
The primary operating mode for laboratory is to allow entering cold water temperatures to float with seasonal
conditions – in order to simulate actual conditions for a home in Syracuse. The plot in Figure 1 shows the typical
annual profile of cold water temperatures from a Syracuse home; conditions at laboratory facility were found be
similar. However, in order to complete a standardized test according to the DOE Test Procedure, we also added a
mixing valve and buffering tank at the cold water inlet to the test setup. This allowed for some heated water from
the tank to be blended in with the cold water stream to achieve the 58°F required by the DOE Test Procedure. This
approach of blending in hot water allowed us to run a standardized test for about half the year (i.e., in the winter
when the city water temperature is below 58°F).
12
Figure 1. Annual Trend of Cold Water Inlet Temperature in the City of Syracuse
Cold Water Inlet - Syracuse Home
80
Temperature (F)
70
60
58°F required for DOE test procedure
50
40
30
January
March April
May
June
July
August
October
2010
2011
Instrumentation was installed to measure:


The energy content of the delivered hot water from the system,
The fuel and electric input into the water heating system.
From these measurements the overall daily efficiency can be determined for various seasons. We also track various
environmental and operating conditions such as:





temperature and relative humidity in the space (or entering the heat pump water heater), and
solar flux striking the collector surface,
flue gas temperature for combustion appliances,
temperature stratification in one of the hot water tanks,
hot water temperature leaving each system.
The laboratory setup has eight (8) separate stations to test nine (9) DHW systems. In Section 2 (Table 7) we
identified the hot water heaters that were included in this test.
13
3.1.1 Laboratory Setup
The laboratory was constructed at the Syracuse Center of Excellence (COE) in two locations. Two (2) gas-fired
units are located in stations 1 and 2 on the first floor. The 1st floor location allows for a conventional natural-draft,
gas-fired water heater, by using the existing flue in the space. The other six stations are located on the fourth floor.
Three of the test stations allow for power-vented gas-fired units (stations 5, 6 and 7) and three other stations
(stations 3, 4, and 8) can support electric units, including the conventional electric tank, two solar systems, and a
heat pump water heater. The solar collectors are located on the roof of the building, which is above the 5th floor
(see Figure 3). The tank-to-collector height is about 32 ft.
3.1.2 Instrumentation and Measurements
Figure 2 schematically shows the location of the instrumentation for each water heating system. Each data point has
a suffix of 1 through 8 corresponding to the station or system number. The environmental conditions are designated
as “A” for the first floor and a “B” for the fourth floor. All the monitored points are listed in Table 8.
For each system, the cold inlet temperature (TC), hot outlet temperature (TH), and water flow (FW) are measured
to determine delivered energy (QH) from the water heater. The temperature sensors are low-mass thermocouples
inserted in the piping in order to quickly respond to changes in temperature with transient flows.
The total unit power use (WE) was measured for each electric unit. Units with a backup resistance heating element
such as the solar systems and heat pump water heater had an additional power measurement (WRE). Gas use (FG)
was measured for the gas units with a temperature-compensated gas meter (0.25 CF per pulse).
The solar systems include measurements for exchanger inlet and outlet temperature (TEI, TEO) as well as the solar
flux (IW5) striking the plane of the collector panel on the roof (see Figure 3). On one of the storage tanks on the
fourth floor, an extended TC probe with 6 sensing points that can be inserted in the tank to measure the degree of
stratification from top to bottom (TTa-TTf). The space conditions are separately measured at the first and fourth
floor locations (TAIA, TAIB). On the fourth floor, relative humidity (RHIB) is measured for the heat pump water
heater. The flue gas temperature (TFG) is measured for all the gas-fired units.
Each test station is built around a metal and plywood frame for mounting the necessary utilities (cold water, hot
water, gas, and electric) and as well as control wiring and instrumentation. A photo of the laboratory test setup is
shown in Figure 4.
14
Figure 2. Schematic Showing Sensor Locations
Hot out
Cold in
TC
TH
Flow Limiters
(1 & 2 gpm)
TFG
Solar
Systems
Only
FW
Solenoid
Valves
TEI, TEO
Space Conditions
WRE
TAIA
TAIB
RHIB
TTa - TTf
WE
Electric
Natural gas
FG
DHW Tank
or Unit
Figure 3. Schematic Showing Solar System on 4th Floor and Roof
IW3
Solar Collectors
Roof
5th Floor
Solar Tank
th
4 Floor
15
Table 8. List of Monitored Points and Instrumentation for Laboratory Testing
System:
Eng
Measurement
Units
Water Inlet Temperature
°F
Water Outlet Temperature
°F
Water Storage Temperature (6 pts) °F
Water Use
gal
Flue Gas/Vent Temperature
°F
Electric Energy Use
kWh
Gas Use
CF
Discharge Air Temperature
°F
Exchanger Inlet Temperature
°F
Exchanger Outlet Temperature
°F
Resistance Element Energy Use
kWh
Solar Flux
W/m2
Space Relative Humidity
% RH
Space Temperature
°F
Outdoor Temperature
°F
Instrumentation Station No:
Watlow type-T 1/16" probe
Watlow type-T 1/16" probe
Watlow multi-pt TC
Omega FTB4605 1/2"
Watlow type-T 1/16" probe
Wattnode MB WNB-3D-240-P
Domestic Meter (0.25 CF/p)
Watlow type-T 1/16" probe
Watlow type-T 1/16" probe
Watlow type-T 1/16" probe
Wattnode MB WNB-3D-240-P
Licor LI200x
Vaisala HMD 60U
Watlow type-T 1/16" probe
CoE Weather Station
GAS-HEPVNT
GAS-STD
SOLARDRAIN
SOLAR /
ELECT-STD
TNKLSCOND
GAS-HECOND
HYBRID
1
2
3
4
5
6
7
8
TC1
TH1
TC2
TH2
TC3
TH3
TC5
TH5
TC6
TH6
TC7
TH7
TC8
TH8
FW1
TFG1
FW2
TFG2
WE2
FG2
FW3
TC4
TH4
TTa to TTf
FW4
WE4
FW6
TFG6
WE6
FG6
FW7
TFG7
WE7
FG7
FW8
WE3
FW5
TFG5
WE5
FG5
TEI3
TEO3
WRE3
IW3
TEI4
TEO4
WRE4
FG1
WE8
TDIS8
WRE8
RHIB
TAIB
TAIA
TAO
Figure 4. Photo of Laboratory Test Setup on Fourth Floor of Syracuse COE
3.1.3 Data Analysis
As described above, the thermal energy delivered to the water heater is calculated in the datalogger at the scan
interval using the formula below:
𝑄𝐻 = 𝑘 ∙ 𝐹𝑊(𝑇𝐻 − 𝑇𝐶)
where
HPWH
k - the product of density and specific heat (8.34 lb/gal∙Btu/(lb∙℉))
16
Water consumption (FW) is measured in gallons by the datalogger. The calculation for QH (in Btu) is completed in
the datalogger for each scan interval (3 seconds). Then the calculated values are summed over each logging interval
to find the total Btu (e.g., per 15 minute interval). The low mass thermocouples are directly inserted in water to
ensure a rapid response to temperature changes that occur during a flow event.
The electric or fuel input is used to determine the total efficiency of the system. This calculation only has meaning
over longer intervals – such as a day – since the energy associated with a water draw may not correspond in time to
the subsequent fuel or power consumption as the water heater recovers. Fuel consumption is based on the higher
heating value of the fuel. Therefore the conversion efficiency for a period is defined as:
𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 𝐻𝐻𝑉
𝑄𝐻
𝑔𝑎𝑠 ∙𝐹𝐺+3412∙𝑊𝐸
where:
QH - Heat Delivery (Btu)
WE - Total Power Use (kWh)
FG - Natural Gas Use (CF)
HHV - Higher heating value for natural gas (1030 Btu per CF )
On units with multiple types of electric use (e.g., the heat pump water heater uses electricity for both the compressor
and the resistance element) the efficiency can be calculated separately for each case. Similarly energy use and hours
in each mode can be tracked.
3.2
Field Testing Approach
One goal of the field testing is to measure the magnitude and timing of hot water loads in several New York State
homes. At all the test sites the hot water use was measured on different time scales to understand the pattern of hot
water use.
Another goal of field test project is to measure the detailed performance of new DHW systems installed in an actual
residence (i.e., at Sites 1, 2 and 3). Sensors were installed to quantify the energy input to these systems as well as
the thermal energy outputs. Intermediate temperatures and component statuses were also tracked to understand
system operation and diagnose any observed anomalies or shortfalls in performance.
In addition, sensors were added at Sites 1 through 5 to determine the specific fixture or appliance associated with
each particular hot water draw. Temperature sensors were installed to track the path (or paths) of each water draw.
This testing provided crucial data to help understand the detailed makeup of water draws in a typical Northeastern
home. This data helped estimate the magnitude of the distribution losses associated with a piping system by
measuring the time delays for given water draw to reach a fixture. This effort to measure, analyze and understand
distribution losses was jointly funded by NYSERDA and the US DOE’s Building America (BA) program (via the
17
ARIES Collaborative). The test efforts to measure distribution losses considered typical homes as well as homes
where new systems and/or distribution and piping improvements were implemented.
A total of 18 sites were included in this field test. Table 9 summarizes the type of retrofit and testing that was
completed at each field site.
Table 9. Summary of Testing at the Field Test Sites
No
NYSERDA
Site ID
Building
America (BA)
Designation
Site/System Changes
Monitoring Approach
1
Detailed #1
BA Group #3
“full retrofit”
A new DHW system installed and
distribution system improvements
implemented.
Monitoring both before and
after changes/retrofit
2
Detailed #2
BA Group #3
“full retrofit”
A new DHW system installed and
distribution system improvements
implemented.
Monitoring both before and
after changes/retrofit
3
Simple #1
BA Group #2
“New Tankless”
A new tankless unit installed (no
distribution changes)
Monitoring both before and
after changes/retrofit
4
Simple #2
BA Group #1
“existing”
No changes
Monitor existing system
5
Simple #3
BA Group #1
“existing”
No changes
Monitor existing system
6
to
18
Simple #4
to #16
-
No changes
Monitor existing system
3.2.1 Instrumentation and Measurements at Field Test Sites
Water-Use Only Sites
At the 13 sites (Sites 6 through 18) without additional BA monitoring, a flow meter was installed to measure the hot
water flow rate. The hot water flow rate was measured with a high resolution Omega flow meter installed on the
cold water inlet of the DHW system. The data were collected at 15-minute intervals using a Campbell Scientific
datalogger. At 8 of these 13 sites, we used a Campbell Scientific CR200x datalogger that additionally collected
flow data at 5-second intervals around each draw event. Short time step data collection was initiated once any flow
occurs (i.e., when one or more pulses are detected) and then logging continued for at least 30-seconds after water
flow subsides. At the other 5 sites, a Campbell Scientific CR10 or CR10X logger (with less memory capacity) was
used to collect only 15-minute data. The data was manually collected (by visiting each site) every 2 to 3 months.
For the dataloggers collecting short time step data, the datalogger memory was partitioned so that a continuous,
uninterrupted stream of 15-minute data was always collected. The data were loaded into a database at CDH,
checked for the validity, and summarized/analyzed in PDF reports. The reports for each site were updated and
posted on the project website. At least 9 months of data was collected from each site.
18
Detailed / BA Sites
At the five BA sites in Table 9 above (including the two sites designated as “NYSERDA Detailed Field” sites), a
monitoring system was installed to measure the detailed performance of the water heating system as well as the hot
water distribution system. Instrumentation was installed to measure:






The energy content of the delivered hot water from the unit,
The fuel and/or electric input into the water heating unit (only Sites 1 through 3),
The environmental conditions (air temperature) near the water heater temperature and piping; the relative
humidity was measured if the system is a heat pump water heater.
Flue gas temperature for combustion appliances,
The runtime or status of each component (e.g., resistance elements, fans, pumps, etc.)
Key temperatures in the trunks and branches leading to each fixture
A Campbell Scientific CR1000 datalogger was used to collect averaged or totalized data at 15-minute intervals for
several months. The datalogger was programmed to sample each sensor at a scan rate of 5 seconds. Then several
key data points were recorded at 5-second intervals when a draw is occurring (i.e., when one or more pulses were
detected). Short time step data collection continued for 30 seconds after the water flow subsided. The collected data
points required for the detailed sites are listed in Table 10. Figure 5 schematically shows the location of the
instrumentation for each water heating system. The specific monitoring points installed at each site are given in
Appendix A.
For each system, the cold inlet temperature (TC), hot outlet temperature (TH), and water flow (FW) are measured
to determine delivered energy (QH) from the water heater. The type-T thermocouples were attached to the outside
of the copper pipe using thermally conductive paste and were well insulated to shield it from ambient conditions
surrounding the pipe.
The total unit power use (WE) was measured for each electric unit. Units with a resistance heating element such as
electric tanks or a heat pump water heater had a status measurement to record the runtime of that component (SE1,
SE2). Gas use (FG) was measured for the gas units with a temperature-compensated gas meter (0.25 CF per pulse).
The flue gas temperature (TFG) was measured for all the gas-fired units. Ambient air temperatures around the
water heater and in unconditioned areas with significant hot water piping runs were also measured. (TA1, TA2,
etc.). Relative humidity (RHI) was measured at sites with a heat pump water heater.
We also installed low mass thermocouples on the outside of the trunk, branch and fixture pipes (i.e., “twigs”) to
determine the path of hot water flow during each draw. Where practical, we installed a thermocouple on the
dedicated line or twig to each point of hot water use or fixture (TF1, TF2, etc.). Temperature sensors were installed
on trunk or branch lines to major areas (kitchen, bath, etc.) to help determine the path of hot water flow (TT1, TT2,
etc.). We used a wireless Campbell Scientific CR206X datalogger to take the temperature measurements in areas
19
that could not be reached with hard-wired sensors. Due to their higher cost, we purchased five of these wireless
loggers to extend monitoring at the five detailed sites. We used two of these wireless loggers at Site 1, two at Site 4,
and one at Site 5.
The location of each fixture was surveyed to establish the effective pipe lengths and volume of the distribution
system (see Appendix A).
Figure 5. Schematic Showing Water Heater System and Distribution Piping
sink
dishwasher
TF2
TF1
sink
Kitchen
TF3
bath
TF4 Bath 1
TF5
TA2
TT3
TT2
TF6
TT4
TT1
TF7
TEX
exhaust
gas
TF8
Utility
sink
Washing
Machine
TH
TA1
SE1
SE2
DHW
Unit
WE
electric
TC
FW
sink
Trunk
FG
natural
gas
20
bath
Bath 2
Table 10. List of Monitored Points and Instrumentation
Detailed
BA3
Detailed
BA3
Simple
BA2
Simple
BA1
Simple
BA1
Simple
1
2
3
4
5
6 to 18
Watlow type-T TC 1/16" probe
TC1
TC2
TC3
TC4
TC5
°F
Watlow type-T TC 1/16" probe
TH1
TH2
TH3
TH4
TH5
gal
Omega FTB4605 1/2"
FW1
FW2
FW3
FW4
FW5
°F
Watlow type-T TC Air
TFG1
TFG2
TFG3
TFG4
TFG5
5-sec
Data ?
Eng
Units
Inlet Water Temperature
Y
°F
Hot Water Oulet Temperature
Y
Hot Water Use
Y
Measurement
Flue Gas/Vent Temperature
Electric Energy Use
Instrumentation
kWh Wattnode MB WNB-3Y-208-P
WE for applicable units
Unit Gas Use
CF
Domestic Meter Pulser (0.25CF/p)
Relative Humidity
%
Vaisala HMD60
Space Temperature near Unit
°F
Watlow type-T TC 1/16" probe
TA1-1
TA1-2
TA1-3
TA1-4
TA1-5
Space Temperature near piping
°F
Watlow type-T TC 1/16" probe
TA2-1
TA2-2
TA2-3
TA2-4
TA2-5
FG1
FG2
FG2
RH (only for HPWH site)
Trunk Temperature - Location 1
Y
°F
Watlow type-T TC 1/16" probe
TT1-1
TT1-2
TT1-3
TT1-4
TT1-5
Trunk Temperature - Location "n"
Y
°F
Watlow type-T TC 1/16" probe
TTn-1
TTn-2
TTn-3
TTn-4
TTn-5
"Twig" Temperature - Fixture 1
Y
°F
Watlow type-T TC 1/16" probe
TF1-1
TF1-2
TF1-3
TF1-4
TF1-5
"Twig" Temperature - Fixture "n"
Y
°F
Watlow type-T TC 1/16" probe
TFn-1
TFn-2
TFn-3
TFn-4
TFn-5
°F
CoE Weather Station
Outdoor Temperature
FW6 - FW18
TAO
The first five test sites used CR1000 dataloggers that were connected to the wireless network in each home via a
wireless bridge. The dataloggers were programmed to send data to the CDH servers each night. The data was
verified and then sent to a custom website that summarizes the performance of each system.
3.2.2 Data Analysis Procedures
The thermal energy delivered by the water heater was calculated in the datalogger at the scan interval using the same
formulas described in the laboratory section above. The total conversion efficiency is also defined as the thermal
delivered energy divided by the fuel and power inputs, all expressed in the same engineering units. Conversion
efficiency is most meaningful on a daily, monthly or annual time scale.
For each hot water draw event we detected the associated end use fixture using the temperature data. We developed
automated analysis procedures similar to Barley et al (2010) in order to detect the fate of each draw by detecting a
change in temperature at each fixture and trunk line. We developed a logic table for each site to assign draws based
on the trunk and branch sensor(s) associated with each fixture. The time delay in hot water reaching each fixture
was used to estimate the volume of water that is “wasted” during each draw event. Hot water waste is defined as
21
water leaving the fixture at less than 100°F or 105°F as suggested by Klein and Barley3. When water draws
simultaneously occurred from multiple fixtures, we had hoped to develop procedures to apportion each draw volume
to the active end uses. However, this provided to be impractical. Figure 6 shows a typical draw profile starting with
cold pipes.
Figure 6. Profile for a Typical Water Draw Event (from Klein 2011)
In most cases it was possible to install a temperature sensor near the end use fixture; however, was still be possible
to infer the water waste associated with each draw by installing a temperature sensor at an intermediate location
(e.g., where a hot water line leaves the basement, heading towards a fixture).
The Building America Report (Henderson and Wade 2014) fully presents the results using these analysis techniques.
Section 5.5 in this report provides a summary of those results.
3
Hot water is only deemed “useful” once it reaches this temperature threshold
22
4
Laboratory Test Results
4.1
Comparing Measured Laboratory Performance to Energy Factor
The first testing conducted in the laboratory was to duplicate the current DOE Test Procedure, as defined in
10CFR430 Subpart B, Appendix E and in ASHRAE Standard 118.2. This procedure consists of six draws one hour
apart. Each draw is 10.7 gallons at 3 gpm for a daily total of 64.2 gallons.
After the initial shakedown and verification in March and early April 2013, the DHW systems started to operate on
April 10, 2013 with this standard draw profile used in the DOE Test Procedure. Figure 7 and Figure 9 shows that
there were still variations in the water draws and total HW use per day that were caused by problems with the flow
limiters. Table 11 shows that the average water use for each system ranged from 60 to 75 gallons per day.
Figure 7. Water Draw Profile Imposed on Water Heaters – CURRENT DOE Draw Profile
04/11/13
14
Water Use: 72.4 gal/day
12
HW Use (gal)
10
8
6
4
2
0
22:
0:
10
11
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
12
Figure 8 and Table 11 summarize the conversion efficiencies measured for each system and compare them to the
rated EF. The standard efficiency gas tank (Gas-STD) and the high-efficiency power vented unit (HE-PVNT) are
both in very good agreement with the measured EF. The conversion efficiency for the tankless water heater
(TANKLESS) is also within 5% of the rated EF. The measured efficiency for the heat pump water heater (HPWH)
actually exceeds the rated value by 5% because the space conditions were slightly warmer than the rated conditions
of 68°F and 50% RH. This standard water use profile did not induce any resistance element operation for the
HPWH. The solar system (SOLAR) had an effective conversion efficiency of 2.3 for this spring period, which is
23
lower than the SEF from the OG-300 rating from SRCC. The glycol-filled drain back system (SOLAR-DRAIN)
was not properly commissioned in this period so it did not function as expected (see Section 4.5).
The tankless unit with a small tank (HYBRID) locked out whenever the HPWH ran. This problem was fixed around
April 26, but the overall efficiency only improved slightly. The HYBRID system efficiency was still well below
expectations.
The condensing tank had an efficiency about 10% below the rated thermal efficiency (TE). Field testing by the
manufacturer had similarly shown applied efficiencies of 0.81 for earlier versions of this unit (Adams 2008). Field
testing in California (Hoeschele and Brand 2013) also showed that efficiency for this unit was about 80% of the TE.
Canadian field testing (Neale 2013) also measured efficiencies of 0.83 for this unit.
Table 11. Summary of Water Heater Performance – CURRENT DOE Draw Profile
April 10 to May 20, 2013
Rated
Efficiency
Ratio
HW Use
(gal/day)
Supply
Temp (F)
Inlet
Temp
(F)
Name
Conversion
Efficiency (-)
Gas STD
0.58
0.58
EF
99.8%
71.1
135.8
52.8
HE-PVNT
0.68
0.67
EF
100.9%
71.6
131.9
49.5
SOLAR
2.32
3.20
SEF
72.4%
75.3
132.0
55.6
SOLAR-DRAIN
1.11
61.0
116.6
54.4
TANKLESS
0.90
0.94
EF
95.2%
61.0
117.6
59.9
HE-Cond
0.86
0.96
TE
89.5%
64.7
134.9
58.9
HYBRID
0.70
0.90
TE
77.6%
62.3
133.5
71.2
HPWH
2.53
2.40
EF
105.3%
59.7
128.0
57.3
Notes: The solar system has a Solar Energy Factor (SEF) instead of an Energy Factor (EF). Similarly, the High
Efficiency Condensing Tank (HE-Cond) and the HYBRID unit (with a small tank) are rated as commercial units with a
thermal efficiency (TE) instead of an EF.
The test conditions were not fully consistent with those specified in the DOE Test Procedure, but were usually close.
For instance the inlet water temperature is specified as 58°F in the procedure. The actual lab test conditions, shown
in Figure 10 and Table 11, vary slightly from that specified value. Similarly, the supply temperature from the water
heater is specified in the procedure as 135°F. Most systems were near that point for this test with the exception of
the SOLAR-DRAIN and the TANKLESS units.
24
Figure 8. Comparing Measured Conversion Efficiency and Energy Factors – CURRENT Draw
Profile
04/10/13 to 05/20/13
6
Conversion Efficiency (-)
5
4
SEF=3.2
SEF=3.1
3
EF=2.4
2
1
8
April
Gas STD
15
22
6
29
May
Solar Drain Tankless
HE-PVNT Solar
HE-Cond
20
13
Hybrid
HPWH
04/10/13 to 05/20/13
1.0
TE=0.96
EF=0.94
Conversion Efficiency (-)
TE=0.90
0.8
EF=0.67
0.6
EF=0.58
0.4
8
April
Gas STD
15
HE-PVNT Solar
22
29
6
May
Solar Drain Tankless
HE-Cond
25
13
Hybrid
20
HPWH
Figure 9. Daily Hot Water Use and Supply Temperatures – CURRENT Draw Profile
04/10/13 to 05/20/13
HW Use (gal/day)
80
60
40
8
15
22
29
April
6
13
20
6
13
20
May
Supply Temperature (F)
04/10/13 to 05/20/13
140
120
100
8
15
April
Gas STD
22
HE-PVNT Solar
29
May
Solar Drain Tankless
HE-Cond
Hybrid
HPWH
Figure 10. Useful Energy vs. Inlet Water Temperature – CURRENT Draw Profile
04/10/13 to 05/20/13
60
Useful Energy (MBtu/day)
50
40
30
20
55
Gas STD
60
HE-PVNT Solar
65
70
Inlet Temperature (F)
Solar Drain Tankless HE-Cond
26
Hybrid
75
HPWH
4.2
Impact of NEW Proposed Hot Water Use Profiles
The proposed ASHRAE Standard 118.2 test procedure proposed three more realistic HW usage profiles to represent
low, normal, and high use (ARHI 2013). Each profile has 12 events that are spread across the 24 hour period with
most events in the morning or evening. Table 12 lists the events in these three profiles along with the total daily
use4.
Table 12. Summary New Hot Water Use Profiles for Proposed ASHRAE 118.2 Test Procedure
Normal Use
High Use
Low Use
Time
Rate
(gpm)
Time
(minutes)
Volume
(gallons)
Rate
(gpm)
Time
(minutes)
Volume
(gallons)
Rate
(gpm)
Time
(minutes)
Volume
(gallons)
6:00
1
2
2
1
2
2
1
2
2
6:30
1.7
9
15.3
3.4
8
27.2
1.5
9
13.5
7:40
1.7
8
13.6
1.7
9
15.3
1
1
1
16:30
1
4
4
1
4
4
1
4
4
16:45
1
3
3
1
3
3
1
3
3
17:30
1
3
3
1.7
2.5
4.25
1
3
3
18:00
1
1
1
1.7
9
15.3
1
1
1
18:45
1
1
1
1
1
1
1
1
1
18:50
1
1
1
1
1
1
1.5
1
1.5
22:15
1.7
8
13.6
1
2
2
1
2
2
22:45
1.7
1
1.7
1.7
1
1.7
1
2
2
23:00
1
5
5
1
6
6
1.5
4
6
64.2
82.75
40
The specified draw rates of 1, 1.7 and 3.4 gpm in the table are different than the capabilities of the laboratory test
setup, which was designed to provide rates of 1, 2 or 3 gpm. To further complicate matters the flow limiters
installed on each test station provided a slightly different flow than the nominal value. For example the 1 gpm flow
limited allowed from 1 to 1.4 gpm of actual flow depending on the flow station, while the 2 gpm limiters provided
1.7 to 2.2 gpm of actual flow. Correction factors were applied to each station to determine the required solenoid
runtime to achieve the desired total volume for each event.
4
The final test procedure issued by DOE in July 11, 2014 actually used slightly different water use profiles then the proposed
ASHRAE 118.2 profiles from March 2013, which are included in Table 12. The final low, medium and high use
profiles had total water use of 38, 55, and 84 gallons per day with 11, 12, draw events, respectively. The final test
procedure is available at http://www.regulations.gov/#!docketDetail;D=EERE-2011-BT-TP-0042.
27
4.2.1 Proposed Normal Use Draw Profile
The proposed “Normal Use” pattern is a more realistic draw profile made up of 12 events with a total water use of
64.2 gallons. The resulting draw profile for Unit 1 is shown in Figure 11 below (the variation from the desired 64.2
gallons per day was due to flow limiter variations). This profile was imposed on all the systems starting in
September 18, 2013. The supply temperature was also lowered to 120°F on some units in order to reflect more
realistic use patterns (except the SOLAR system, which remained at 135°F).
Figure 11. Water Draw Profile Imposed on Water Heaters – Normal Use Profile
09/19/13
20
Water Use: 65.2 gal/day
HW Use (gal)
15
10
5
0
22:
0:
18
19
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
20
Table 13 and Figure 12 show the measured conversion efficiency for these slightly different operating conditions.
Similar overall performance was observed with this new use pattern compared to the CURRENT profile. The
conversion efficiency for the solar systems are slightly lower since the weather was slightly cloudier in this early fall
test period. The lockout problem with the HYBRID system was corrected in this period so the efficiency is slightly
higher – but still is much lower than the value implied by the thermal efficiency. This water use profile also did not
induce any resistance element operation for the HPWH unit.
28
Table 13. Summary of Water Heater Performance – Normal Use Profile
September 18 to October 31, 2013
Rated
Efficiency
Ratio
HW Use
(gal/day)
Supply
Temp (F)
Inlet
Temp
(F)
Name
Conversion
Efficiency (-)
Gas STD
0.56
0.58
EF
96.7%
64.90
119.5
65.9
HE-PVNT
0.66
0.67
EF
98.8%
81.80
125.2
65.5
SOLAR
1.90
3.20
SEF
60.90
130.7
68.1
SOLAR-DRAIN
1.39
63.80
117.4
66.9
TANKLESS
0.86
0.94
EF
91.3%
67.00
119.5
67.1
HE-Cond
0.85
0.96
TE
88.9%
64.70
122.7
67.8
HYBRID
0.73
0.90
TE
81.3%
60.30
114.8
66.0
HPWH
2.70
2.40
EF
112.5%
62.20
120.0
67.7
Notes: The solar system has a Solar Energy Factor (SEF) instead of an Energy Factor (EF). Similarly, the High
Efficiency Condensing Tank (HE-Cond) and the HYBRID unit (with a small tank) are rated as commercial units with a
thermal efficiency (TE) instead of an EF.
29
Figure 12. Comparing Measured Conversion Efficiency and Expected Energy Factors – Normal
Use Profile
09/18/13 to 10/31/13
6
Conversion Efficiency (-)
5
4
SEF=3.2
SEF=3.1
3
EF=2.4
2
1
16
23
September
Gas STD HE-PVNT Solar
30
7
14
October
Solar Drain Tankless
HE-Cond
21
Hybrid
28
HPWH
09/18/13 to 10/31/13
1.0
TE=0.96
EF=0.94
Conversion Efficiency (-)
TE=0.90
0.8
EF=0.67
0.6
EF=0.58
0.4
16
23
September
Gas STD HE-PVNT Solar
30
7
14
October
Solar Drain Tankless
HE-Cond
30
21
Hybrid
28
HPWH
Figure 13 shows the daily variations as well as the supply temperatures for each system.
Figure 13. Daily Hot Water Use and Supply Temperatures – Normal Use Profile
09/18/13 to 10/31/13
HW Use (gal/day)
80
60
40
16
23
September
30
7
14
21
28
21
28
October
Supply Temperature (F)
09/18/13 to 10/31/13
140
120
100
16
23
September
Gas STD HE-PVNT Solar
30
7
14
October
Solar Drain Tankless
HE-Cond
Hybrid
HPWH
4.2.2 Proposed Low Use Draw Profile
The proposed “Low Use” pattern is a more a realistic draw profile for a small household made up of 12 events with
a total water use of 40 gallons. The resulting profile for one of the stations is shown in Figure 14 below (the slightly
different use for this day was due to flow limiter variations). This profile was imposed on several systems starting
on June 18, 2014. Operation of standard water heater (Gas-STD) was stopped for this test due to problems keeping
the pilot lit. Unreliable solenoid operation also caused us to stop operation on Test Station 2 (HE-PVNT) and Test
Station 6 (HE-Cond). The supply temperature was set near 120°F for all systems except the SOLAR-DRAIN as
shown in Figure 16. This water use profile did not induce any resistance element operation for the HPWH unit.
31
Figure 14. Water Draw Profile Imposed on Water Heaters – Low Use Profile
Unit 5 07/05/14
15
Water Use: 44.3 gal/day
HW Use (gal)
10
5
0
22:
0:
4
5
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
6
Table 14 and Figure 15 show the measured conversion efficiency for with this Low Use profile. Similar overall
performance was observed with this new use pattern. The SOLAR system conversion efficiency is higher for this
summer period (as expected). Several units were not operating due to solenoid problems. The efficiency of the
HYBRID system and the TANKLESS unit is still much lower than the value implied by the thermal efficiency or
energy factor. The EF for HPWH is very close to rated performance.
Table 14. Summary of Water Heater Performance – Low Use Profile
July 4 to September 22, 2014
Name
Conversion
Efficiency (-)
Rated
Efficiency
Gas STD
0.58
EF
HE-PVNT
0.67
EF
3.20
SEF
SOLAR
794
SOLAR-DRAIN
117
TANKLESS
0.81
HE-Cond
0.94
EF
0.96
TE
HW Use
(gal/day)
Supply
Temp (F)
Inlet
Temp
(F)
39.90
123.7
71.1
41.30
118.0
70.4
86.2%
39.30
116.6
71.2
Ratio
HYBRID
0.65
0.90
TE
72.2%
41.90
115.3
71.2
HPWH
2.46
2.40
EF
102.5%
41.70
117.3
70.7
Notes: The solar system has a Solar Energy Factor (SEF) instead of an Energy Factor (EF). Similarly, the High
Efficiency Condensing Tank (HE-Cond) and the HYBRID unit (with a small tank) are rated as commercial units with a
thermal efficiency (TE) instead of an EF.
32
Figure 15. Comparison of Measured Conversion Efficiency and Expected Energy Factors – Low
Use Draw Profile
07/04/14 to 09/22/14
6
Conversion Efficiency (-)
5
4
SEF=3.2
SEF=3.1
3
EF=2.4
2
1
30 7
July
Gas STD
14
21
28
4
11
HE-PVNT Solar
25
18
August
Solar Drain Tankless
1
8
September
HE-Cond Hybrid
15
22
HPWH
07/04/14 to 09/22/14
1.0
TE=0.96
EF=0.94
Conversion Efficiency (-)
TE=0.90
0.8
EF=0.67
0.6
EF=0.58
0.4
30 7
July
Gas STD
14
21
HE-PVNT Solar
28
4
11
18
August
Solar Drain Tankless
33
25
1
8
September
HE-Cond Hybrid
15
HPWH
22
Figure 16. Daily Hot Water Use and Supply Temperatures – Low Use Profile
07/04/14 to 09/22/14
HW Use (gal/day)
50
45
40
35
30
30 7
14
21
28
July
4
11
18
25
August
1
8
15
22
15
22
September
Supply Temperature (F)
07/04/14 to 09/22/14
140
120
100
30 7
July
Gas STD
14
21
HE-PVNT Solar
28
4
11
18
August
Solar Drain Tankless
25
1
8
September
HE-Cond Hybrid
HPWH
4.2.3 Proposed High Use Draw Profile
The proposed High Use pattern is a more realistic profile for a large family made up of 12 events with a total water
use of 82.8 gallons per day. The resulting profile for one of the stations is shown in Figure 17 below (the slightly
different use for on this day was due to flow limiter variations). This profile was imposed on several systems
starting on January 15, 2015. The supply temperature was set near 120°F for all systems. Not all the systems were
operating this point due to flow limiter problems and solenoid failures.
34
Figure 17. Water Draw Profile Imposed on Water Heaters – High Use Profile
Unit 5 01/17/15
30
Water Use: 84.4 gal/day
25
HW Use (gal)
20
15
10
5
0
22:
0:
16
17
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
18
Table 15 and Figure 18 show the measured conversion efficiency for the units with this High Use profile. The
SOLAR system conversion efficiency is lower for this winter period (as expected). Several units were not operating
due to solenoid problems. The efficiency of the HYBRID system and the TANKLESS unit are still lower than the
value implied by the thermal efficiency or energy factor, but are higher than was observed with the lower usage
profiles.
Table 15. Summary of Water Heater Performance – High Use Profile
January 15 to March 29, 2015
Name
Conversion
Efficiency (-)
Gas STD
0.58
EF
HE-PVNT
0.67
EF
3.20
SEF
SOLAR
1.15
SOLAR-DRAIN
1.30
TANKLESS
0.88
HE-Cond
Ratio
HW Use
(gal/day)
Supply
Temp (F)
Inlet
Temp
(F)
36.0%
77.00
119.5
47.9
75.10
118.1
45.0
94.0%
83.30
117.6
52.2
Rated
Efficiency
0.94
EF
0.96
TE
HYBRID
0.78
0.90
TE
86.1%
77.70
113.7
48.8
HPWH
1.77
2.40
EF
73.8%
79.30
114.0
47.2
Notes: The solar system has a Solar Energy Factor (SEF) instead of an Energy Factor (EF). Similarly, the High
Efficiency Condensing Tank (HE-Cond) and the HYBRID unit (with a small tank) are rated as commercial units with a
thermal efficiency (TE) instead of an EF.
35
Figure 18. Comparison of Measured Conversion Efficiency and Expected Energy Factors – High
Use Profile
01/15/15 to 03/29/15
5
Conversion Efficiency (-)
4
SEF=3.2
SEF=3.1
3
EF=2.4
2
1
12
19
26
2
9
16
2
23
9
March
February
January
Solar Drain Tankless HE-Cond
Gas STD HE-PVNT Solar
16
Hybrid
23
HPWH
01/15/15 to 03/29/15
1.0
TE=0.96
EF=0.94
Conversion Efficiency (-)
TE=0.90
0.8
EF=0.67
0.6
EF=0.58
0.4
12
19
26
2
9
16
23
2
9
January
February
March
Gas STD HE-PVNT Solar
Solar Drain Tankless HE-Cond
36
16
Hybrid
23
HPWH
Figure 19. Daily Hot Water Use and Supply Temperatures – High Use Profile
01/15/15 to 03/29/15
HW Use (gal/day)
100
90
80
70
60
12
19
26
January
2
9
16
23
2
February
9
16
23
9
16
23
March
Supply Temperature (F)
01/15/15 to 03/29/15
140
120
100
12
19
26
2
9
16
23
2
January
February
March
Gas STD HE-PVNT Solar
Solar Drain Tankless HE-Cond
Hybrid
HPWH
The conversion efficiency for HPWH is 1.77 compared to the rated EF of 2.4, because this water use profile induced
resistance element operation after the larger draw. The performance of the HPWH unit over one day (February 15,
2015) is shown by Figure 20. The morning draw of 25 gallons induces the resistance elements to come on. The
upper element comes on for 12.5 minutes followed by 40.5 minutes of lower element operation. Note that the
control logic of the unit attempts to recover as quickly as possible, so compressor operation is locked out until the
bottom element in the tank can fully bring the tank back to temperature. Similar performance for the same HPWH
unit was observed in the field as discussed in Section 5.6.
37
Figure 20. Impact of High Use Profile on HPWH Performance
02/12/15
HPWH HW Use (gal)
FW= 79.2 gal/day
25
2.0 13.0
20
25.0
DRAW= 25.0 gal
4.7 2.0
6.516.3
7.7
2.0
15
10
5
0
22:
0:
11
12
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
UPPER= 12.5 min,
LOWER= 40.5 min
ON
LOWER
OFF
ON
UPPER
OFF
22:
0:
11
12
WE8=
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
02/12/15
7.0 kwh/day
5
Power (kW)
4
3
2
1
0
22:
0:
11
12
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
February
2015
38
4.3
Comparing Performance with Different Usage Profiles
The measured conversion efficiencies with the different draw profiles are summarized in Table 16. As mentioned
above, not all the profiles were run for each system due to problems with solenoids and pilot lights. A series of plots
also shows the impact of each usage profile on conversion efficiency for each system:







The standard water heater (Gas-STD) in Figure 21 shows that the conversion efficiency was always in line
with the EF. The data show a slight efficiency increase with water use, as would be expected for a storage
tank.
Similarly, the power vented higher efficiency water heater (HE-PVNT) in Figure 22 shows that the
conversion efficiency was also close to the EF and also showed a slight efficiency increase with water use.
The SOLAR (Figure 23) and SOLAR-DRAIN (Figure 24) systems were more affected by differing
performance on sunny days and cloudy days (Section 4.5 discusses solar performance in greater detail).
The tankless unit in Figure 25 shows that the conversion efficiency always fell short of the 0.94 EF. Some
efficiency improvement with greater water use was apparent. The large draws of the current procedure
results in the best conversion efficiency – implying that the efficiency predicted by the current procedures
overstates actual performance.
The High Efficiency Condensing (HE-Cond) unit in Figure 26 always had conversion efficiencies that fell
short of the TE of 0.96. While tests were not run for all the draw profiles, the efficiency does modestly
increase with flow rate.
The HYBRID unit in Figure 27 also had conversion efficiencies that fell short of the TE of 0.90. This unit
showed more variation with water use than the other systems.
The HPWH unit data in Figure 28 confirm the strong impact that usage profile has on the conversion
efficiency. At the normal and low use profiles the unit runs without resistance element operation and the
conversion efficiency is in good alignment with the EF (the variations are mostly due to changing space
conditions). For the High Use case, the element comes on and reduces the conversion efficiency by 75%.
Table 16. Summary of Conversion Efficiencies with Different Draw Patterns
Name
Rated
Efficiency
Current
Pattern
Normal
Use
Low
Use
High
Use
64.2
gpd
64.2
gpd
40 gpd
82.8
gpd
Gas STD
0.58
EF
0.58
0.56
HE-PVNT
0.67
EF
0.68
0.66
SOLAR
3.20
SEF
2.32
1.90
794
1.15
1.11
1.39
117
1.30
0.81
0.88
SOLAR-DRAIN
TANKLESS
0.94
EF
0.90
0.86
HE-Cond
0.96
TE
0.86
0.85
HYBRID
0.90
TE
0.70
0.73
0.65
0.78
HPWH
2.40
EF
2.53
2.70
2.46
1.77
39
Figure 21. Impact of Draw Profile on Conversion Efficiency – Gas-STD
Gas-STD - EF=0.58
0.8
Conversion Efficiency (-)
Current Pattern
Normal Use
Low Use
High Use
Rating
0.6
0.4
20
40
60
80
100
Hot Water Use (gal/day)
Figure 22. Impact of Draw Profile on Conversion Efficiency – HE-PVNT (other)
HE-PVNT - EF=0.67
0.8
Conversion Efficiency (-)
Current Pattern
Normal Use
Low Use
High Use
Rating
0.6
0.4
20
40
60
80
Hot Water Use (gal/day)
40
100
Figure 23. Impact of Draw Profile on Conversion Efficiency – SOLAR
Solar - SEF=3.2
3.0
Conversion Efficiency (-)
2.5
2.0
Current Pattern
Normal Use
Low Use
High Use
1.5
1.0
0.5
20
40
60
80
100
Hot Water Use (gal/day)
Figure 24. Impact of Draw Profile on Conversion Efficiency – SOLAR-DRAIN
Solar Drain 3.0
Conversion Efficiency (-)
2.5
Current Pattern
Normal Use
Low Use
High Use
2.0
1.5
1.0
0.5
20
40
60
80
Hot Water Use (gal/day)
41
100
Figure 25. Impact of Draw Profile on Conversion Efficiency – TANKLESS
Tankless - EF=0.94
1.0
Conversion Efficiency (-)
Rating
Current Pattern
Normal Use
Low Use
High Use
0.8
0.6
20
40
60
80
100
Hot Water Use (gal/day)
Figure 26. Impact of Draw Profile on Conversion Efficiency – HE-Cond
HE-Cond - TE=0.96
1.0
Conversion Efficiency (-)
Rating
Current Pattern
Normal Use
Low Use
High Use
Rating
0.8
0.6
20
40
60
80
Hot Water Use (gal/day)
42
100
Figure 27. Impact of Draw Profile on Conversion Efficiency – HYBRID
Hybrid - TE=0.90
Conversion Efficiency (-)
1.0
Rating
0.8
Current Pattern
Normal Use
Low Use
High Use
Rating
0.6
20
40
60
80
100
Hot Water Use (gal/day)
Figure 28. Impact of Draw Profile on Conversion Efficiency – HPWH
HPWH - EF=2.4
Conversion Efficiency (-)
3.0
2.5
Rating
2.0
Current Pattern
Normal Use
Low Use
High Use
1.5
20
40
60
80
Hot Water Use (gal/day)
43
100
4.4
Standby Losses and Parasitic Power
4.4.1 Standby Losses
From May 22 to June 27, 2013 the water heaters in the laboratory ran in standby mode with no hot water use. This
provided an indication of the standby losses from each system. The air temperature on 1st floor (for DHW1 and
DHW2) was 70°F. The air temperature for the other units on the 4 th floor was 75°F. The measured standby losses
are summarized in Table 17. The figures show the variations in losses over the period.
Table 17. Summary of Water Heater Standby Losses
Unit
Gas Use (MBtu/day)
Electric Use (kWh/day)
DHW1 – GAS-STD
~10
-
DHW2 – HE-PVNT
6.3
0.135
DHW6 – HE-COND
6.9
0.228
DHW7 – HYBRID
8.3
Not measured
DHW8 – HPWH
-
0.688
For DHW1 (AO Smith GCVX-50-100), the conventional, natural-draft water heater, the standby losses showed
much more fluctuation than expected, as shown in Figure 29. The variation may indicate significant differences in
the amount of draft induced through the unit on windy and calm days. The average was 23.2 MBtu/day while the
most common minimum value was just over 10 MBtu/day. The hot water set point was 130-135°F.
Figure 29. Daily Standby Gas Use for DHW1-GAS-STD
DHW1
Daily Gas Use (MBtu/day)
80
60
40
AVG = 23.2
20
0
20
May
27
3
10
17
June
2013
44
24
For DHW2-HE-PVNT (Rheem 43VP40SE2), the power vented unit with an EF of 0.67, the average standby losses
were 6.3 MBtu/day Figure 30). The hot water set point as about 130°F.
Figure 30. Daily Standby Gas Use for DHW2-HE-PVNT
DHW2
12
Daily Gas Use (MBtu/day)
10
8
6
4
AVG =
6.3
2
0
20
27
May
3
10
17
24
June
2013
Average gas use is 6.9 MBtu/day with no water use for the DHW6-HE-Cond (AO Smith Vertex), as shown in
Figure 31. Standby losses reported on AHRI certificate are 499 Btu/h, or 12.0 MBtu/day. Tank set point is 135°F.
Figure 31. Daily Standby Gas Use for DHW6_HE-Cond
DHW6
Daily Gas Use (MBtu/day)
15
10
5
AVG =
6.9
0
20
May
27
3
10
17
June
2013
45
24
For the DHW7-HYBRID the tank set point is 130°F. Average gas use is 8.3 MBtu/day with no water use (Figure
32). Standby losses reported on AHRI certificate are 378 Btu/h, or 9.0 MBtu/day.
Figure 32. Daily Standby Gas Use for DHW7-HYBRID
DHW7
12
Daily Gas Use (MBtu/day)
10
8
6
4
AVG =
8.3
2
0
20
27
May
3
10
17
24
June
2013
Standby losses for DHW8-HPWH (Geospring™ GEH50DEEDSC) are 0.69 kWh/day. The hot water set point was
130°F.
Figure 33. Daily Standby Electric Use for DHW8
DHW8
Daily Electric Use (kWh/day)
1.0
0.8
0.6
0.4
AVG = 0.69
0.2
0.0
20
May
27
3
10
17
June
2013
46
24
4.4.2 Parasitic Electric Loads
All the higher efficiency gas-fired systems also use electricity in addition to natural gas. Figure 34 shows the trend
of daily electric consumption versus daily gas consumption for these four systems. As would be expected all the
system consume electricity even with little or no gas use (the y-intercept). Then electricity increases linearly with
gas use. Table 18 summarizes the standby loss term, the slope of the electric-to-gas and lists the power use with a
normal water use pattern (64.2 gal/day). The Stand loss term varies from 0.1 to 0.3 kWhday. The slope is high for
the power-vented unit (HE-PVNT), which makes sense given relatively large fan on that unit. The other higher
efficiency units generally have a smaller slope, except for the HYBRID. All these uses use less than 0.5 kWh per
day under normal HW use profile.
Figure 34. Trends of Electric Consumption versus Natural Gas Input
HE-PVNT
Tankless
0.25
Electric (kWh/day)
Electric (kWh/day)
0.4
0.2
kWh/therm: 0.358 kWho= 0.10
0.0
0.20
0.15
0.10
0.05
kWh/therm: 0.109 kWho= 0.13
0.00
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.10 0.20 0.30 0.40 0.50 0.60
Gas (therm/day)
Gas (therm/day)
HE-Cond
HYBRID
0.50
Electric (kWh/day)
Electric (kWh/day)
0.8
0.6
0.4
0.2
0.0
kWh/therm: 0.141 kWho= 0.23
0.0
0.2
0.4
0.6
0.40
0.30
0.20
0.10
kWh/therm: 0.333 kWho= 0.30
0.00
0.8
0.0
0.2
Gas (therm/day)
0.4
0.6
0.8
Gas (therm/day)
Table 18. Electric Use of Gas-Fired Systems
System
Standby Losses
(kWh/day)
Slope (kWh/therm)
Electricity w/ Normal
Use (kWh/day)
HE-PVNT
0.10
0.358
0.32
Tankless
0.13
0.109
0.17
HE-Cond
0.23
0.141
0.27
HYBRID
0.30
0.333
0.41
47
4.5
Comparing Performance of Solar Systems
The COE laboratory has two different solar systems installed (see Table 19). The Rheem unit is a conventional
system with a charged glycol loop. This type of system is widely used in New York State. The Wagner is a new
German design that is a drain back system using glycol. Drain back systems are widely used in southern climates
because they offer superior thermal performance. However, the application of drain back systems in colder climates
has been limited due to fear of collector freeze damage. This new design uses a glycol fluid with a combined
reservoir/heat exchanger/pump assembly built into the solar storage tank. This simplifies the installation of the
system and provides freeze protection.
Table 19. Solar Systems Installed in the Laboratory
Collector Area
Rheem
(System 3)
6.10 m2
Wagner
(System 4a/4b)
5.22 m2
System Arrangement
Single tank
120 gal with internal
glycol HX and two electric
elements
Two Tank
92 gal glycol drain back
(internal reservoir & HX)
52 gal std electric tank
SRCC Solar Fraction
0.71
0.80
The Wagner system controller has an option to include booster pump to help overcome the static head on a drain
back system where the collector is much higher than the tank. The system was initially installed without a booster
pump. In an effort to improve on poor performance, a booster pump was added on July 3, 2013 in series with the
main pump to help overcome the initial static head. The Wagner controller only ran the booster pump during the
first 4-5 minutes of main pump operation.
Both solar systems included instrumentation to determine key performance parameters. QU is useful heat delivered
to the load (it is integrated every 3 seconds by the data logger). QSL is the standby losses for the 52 gallon electric
tank. Testing for standby losses in February 2014 showed that this tank required 1 kWh per day to maintain
temperature with no hot water use 5. QE is the supplemental electric energy used to heat the tank as well as the
power for the solar pumps (91 Watts for the Rheem; 85 Watts for 1 pump on the Wagner; and 160 Watts for two
pumps after May 9, 2014). From these values delivered solar energy (QS), solar fraction (SF) and COP were
determined using the equations below.
5
The standby losses for the 120 gallon Rheem tank actually approached 2 kWh/day on cloudy days, but smaller losses for the
52 gallon tank were still assumed since this is a more appropriate base case.
48
QU =
k x FW x (TH – TC)
QS =
QU + QSL - QE
COP = QU / QE
SF =
1 - QE / (QU + QSL)
4.5.1 Initial Performance Observations (September and October 2013)
The initial data collected in September and October 2013 showed that the two solar systems behaved as expected on
cloudy days (the Wagner was slightly better). However, the performance of the Wagner system degraded on clear,
sunny days (with high solar flux). Figure 35 and Figure 36 compare the daily performance of the two solar systems
in that period in terms of solar fraction (SF) and electric element/pump operation (QE). Figure 37 directly compares
the solar fraction and the normalized electric input for the two systems and shows that the Wagner has better
performance on less sunny days while the Rheem is better on clear sunny days.
Figure 35. Solar Fraction vs. Solar Flux for Fall 2013
09/19/13 to 10/31/13
1.0
Rheem: 0.38 SF
Wagner: 0.32 SF
Solar Fraction (-)
0.8
0.6
0.4
0.2
0.0
0
2
4
6
Daily Solar Flux (kWh/m^2)
49
8
Figure 36. Daily Electric Power vs. Solar Flux for Fall 2013
09/19/13 to 10/31/13
12
Rheem: 6.8 kWh/day
Wagner: 6.4 kWh/day
Daily Power Input (kWh/day)
10
8
6
4
2
0
0
2
4
6
8
Daily Solar Flux (kWh/m^2)
Figure 37. Direct Comparison of Rheem vs. Wagner Performance for Fall 2013
Electricity
Wagner Power (kWh/gal)
Solar Fraction
Wagner Fraction (-)
1.0
0.8
0.6
0.4
0.2
0.0
0.0
0.2
0.4
0.6
0.8
1.0
Rheem Fraction (-)
COP
Wagner COP (-)
3
50
1
0.15
0.10
0.05
0.00
0.00
0.05
0.10
0.15
Rheem Power (kWh/gal)
4
2
0.20
0.20
The two sets of plots below illustrate the Wagner performance issue by showing detailed performance on two
different days. Figure 38 shows a moderately sunny day (October 1) when the total flux for the day was 3.4
kWh/m2. The fluid temperatures to and from the solar collector are similar for the two systems. Figure 39 shows a
clear, sunny day (September 25) when the total flux for the day was 6.6 kWh/m 2. In this case the collector
temperatures on the Wagner system (as measured near the tank) started the day increasing as expected, but at about
10 am the temperatures suddenly dropped. Temperatures remained low until about 4 pm, even though the pump
remained running the entire time.
This anomaly was due to the pump’s inability to overcome both the high static head AND the vapor pressure that
builds up in the collector as the fluid gets hot. As a result the flow for the drain back collector stops because the
fluid in the return line back to the tank vaporizes and breaks the siphon when fluid gets too hot (i.e., the system
becomes “vapor-locked”).
For this installation, the height from the tank (4th floor) to the collector (roof or 6th floor) was higher than for than
most residential single applications (the height of two commercial stories is 32 ft). The length of 3/8 inch piping
from the tank to the collector also created more pressure drop to overcome.
51
Figure 38. Performance Comparison for a Moderately Sunny Day (October 1, 2013)
10/01/13
Solar Flux (W/m^2)
1200
1000
800
3.4 kWh/day
600
400
200
0
22:
0:
30
1
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
2
Rheem Pump = 4.8 hr/day,
Wagner Pump = 6.6 hr/day
ON
One Pump
Wagner Pump
OFF
ON
Rheem Pump
OFF
22:
0:
30
1
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
2
10/01/13
Rheem Collector Outlet
Rheem Collector Inlet
Wagner Collector Outlet
Wagner Collector Inlet
Rheem Solar Frac: 0.52
Wagner Solar Frac: 0.35
150
100
50
22:
0:
30
1
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
2
52
Figure 39. Performance Comparison for a Sunny Day (September 25, 2013)
09/25/13
Solar Flux (W/m^2)
1200
1000
800
6.6 kWh/day
600
400
200
0
22:
0:
24
25
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
26
Rheem Pump = 7.6 hr/day,
Wagner Pump = 8.1 hr/day
ON
One Pump
Wagner Pump
OFF
ON
Rheem Pump
OFF
22:
0:
24
25
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
26
09/25/13
Rheem Collector Outlet
Rheem Collector Inlet
Wagner Collector Outlet
Wagner Collector Inlet
Rheem Solar Frac: 0.64
Wagner Solar Frac: 0.28
150
100
50
22:
0:
24
25
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
26
53
4.5.2 The Wagner Drain-Back System is Fixed
To solve the problem identified above, we wired the two pumps to operate together on May 9, 2014. This forced the
two pumps to always operate together (previously the second pump had been connected to R2 on the controller and
only ran during the first few minutes of startup). Figure 40 is a daily plot that shows the Wagner System started to
work properly on a sunny day after this fix was implemented. Figure 41 and Figure 42 show the same daily trends
as before, but with different symbols for both before and after the Wagner system was fixed. All data in the plots
below correspond to the hot water use of 68-70 gallons per day for both systems.
54
Figure 40. Performance Comparison for a Sunny Day (May 20, 2014)
05/20/14
Solar Flux (W/m^2)
1200
1000
800
7.2 kWh/day
600
400
200
0
22:
0:
19
20
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
21
Rheem Pump = 8.0 hr/day,
Wagner Pump = 8.0 hr/day
ON
Two Pumps
Wagner Pump
OFF
ON
Rheem Pump
22:
0:
19
20
OFF
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
21
05/20/14
Rheem Collector Outlet
Rheem Collector Inlet
Wagner Collector Outlet
Wagner Collector Inlet
Rheem Solar Frac: 0.70
Wagner Solar Frac: 0.70
150
100
50
22:
0:
19
20
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
21
55
Figure 41. Solar Fraction vs. Solar Flux (before and after Wagner Fix)
03/03/14 to 06/17/14
1.0
Rheem: 0.49 SF
Wagner: 0.42 SF
Wagner: 0.72 SF (fixed)
Solar Fraction (-)
0.8
0.6
0.4
0.2
0.0
0
2
4
6
8
Daily Solar Flux (kWh/m^2)
Figure 42. Daily Electric Power vs. Solar Flux (before and after Wagner Fix)
03/03/14 to 06/17/14
12
Rheem: 5.9 kWh/day
Wagner: 6.8 kWh/day
Wagner: 3.4 kWh/day (fixed)
Daily Power Input (kWh/day)
10
8
6
4
2
0
0
2
4
6
Daily Solar Flux (kWh/m^2)
56
8
Figure 43. Direct Comparison of Rheem vs. Wagner Performance (before and after Wagner Fix)
Electricity
Wagner Power (kWh/gal)
Solar Fraction
Wagner Fraction (-)
1.0
0.8
0.6
0.4
0.2
0.0
0.0
0.2
0.4
0.6
0.8
1.0
0.20
0.15
0.10
0.05
0.00
0.00
Rheem Fraction (-)
0.05
0.10
0.15
0.20
Rheem Power (kWh/gal)
COP
Table 20 4
below summarizes the overall impact of the fixing Wagner System (i.e., operating two pumps together).
Wagner COP (-)
The solar fractions for the two systems were within 0.02, but this difference increased to 0.05 when the second
pump was3added. Similarly, before the fix, the amount of electric use was lower for the Wagner system by 0.2
kWh/day. After the fix, the Wagner was lower by 0.6 kWh/day. The imposed draw profile both before and after the
2
fix was 60 gallons per day.
1
Table 20. Summary Impact of Fixing Wagner System
0
0
1
2
3
4
Rheem System
Wagner
System
Difference
Before Fix
(March 3 to May 9)
0.40
0.42
+0.02
After Fix
(May 9 to June 17)
0.67
0.72
+0.05
Rheem System
Wagner
System
Difference
Before Fix
(March 3 to May 9)
7.0
6.8
-0.2
After Fix
(May 9 to June 17)
4.0
3.4
-0.6
Rheem COP (-)
Solar Fraction (-)
Electric Use (kWh/day)
The analysis above focused on short periods in the spring season with the same water use pattern. Table 21
compares the monthly data for the two solar systems from August 2013 through to February 2015, showing the
monthly values for solar fraction, electric use and hot water use. Some of the more subtle differences discussed in
57
the previous sections did not have a significant impact on the overall monthly numbers. Water use varied across the
period as different water use profiles were imposed on the systems in the laboratory.
Table 21. Monthly and Annual Performance Summary for Solar Systems
Solar Fraction (-)
Electric Use (kWh)
HW Use (gal/day)
Month
Solar Flux
(kWh/m2)
Rheem
Wagner
Rheem
Wagner
Rheem
Wagner
Aug-13
159.7
0.65
0.43
123.9
157.9
66.1
70.7
Sep-13
143.1
0.57
0.35
142.1
175.9
59.6
64.7
Oct-13
90.0
0.29
0.30
239.1
206.6
60.5
63.6
Nov-13
74.1
0.24
0.25
220.7
234.5
54.8
62.7
Dec-13
34.6
0.08
0.13
301.3
291.9
61.2
61.2
Jan-14
68.9
0.13
0.15
316.0
308.4
64.4
62.9
Feb-14
58.5
0.09
0.14
215.5
206.1
43.8
42.7
Mar-14
113.5
0.29
0.26
241.3
252.3
67.4
65.1
Apr-14
133.7
0.48
0.53
182.6
166.3
70.6
68.8
May-14
169.4
0.65
0.71
131.9
111.4
70.0
68.0
Jun-14
160.2
0.75
0.74
86.9
87.6
56.7
56.5
Jul-14
164.1
0.98
0.84
34.8
55.2
40.6
41.9
Aug-14
152.8
0.95
0.82
39.0
58.0
39.9
41.4
Sep-14
143.4
0.91
0.80
43.2
57.9
39.4
40.7
Oct-14
90.9
0.54
0.49
101.4
111.8
40.6
41.3
Nov-14
60.2
0.23
0.25
154.3
157.7
40.1
41.1
Dec-14
32.3
0.08
0.13
199.0
198.7
40.1
41.3
Jan-15
76.0
0.15
0.19
300.1
290.4
61.5
61.2
Feb-15
20.4
0.05
0.11
387.4
359.8
76.8
74.5
0.47
0.46
1,746
1,771
51.1
51.0
2014
The overall annual solar fractions for 2014 were 0.47 for the Rheem and 0.46 for the Wagner, with average water
use of 51 gallons per day. The overall annual electric use for electric elements and the solar pumps was 1,746 kWh
for the Rheem and 1,771 kWh for the Wagner. The Wagner has a better solar fraction in the winter months
primarily because the second 92 gallon tank in this two tank system gains heat from the surrounding room air even
when the solar input to the tank is zero. When we bias the 12 month period to end in February 2015, relative
differences between the two systems were not significantly affected.
58
5
Field Test Results
5.1
Characteristics of Field Test Sites
The domestic hot water (DHW) systems were monitored in eighteen (18) homes with one to five permanent
occupants per household (2.7 occupants on average as of January 2013). At households with variable occupancy,
we were able to identify the actual occupancy down to a monthly or daily basis. Figure 44 shows the monitoring
period for each site.
Figure 44. Duration of Field Monitoring Period at Each Test Site
DHW Monitoring by Site
Jun-12
Sep-12
Dec-12 Mar-13 Jun-13
Sep-13
Dec-13 Mar-14 Jun-14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Sep-14
Dec-14
Previous
Owners
Completed
Monitoring
Figure 45 shows the average monthly occupancy for all 18 sites from December 2012 through December 2014.
Several sites had a consistent household occupancy, as shown by the horizontal lines on the plot. Interviews and
surveys allowed us to pinpoint household occupancy at sites where the number of occupants varied over the course
of the year. There were a handful of sites with college-age occupants, one site where an adult traveled for work on a
weekly basis, and another site that added a new member to the household.
59
Figure 45. Average Monthly Occupancy at Each Test Site
Average Monthly Occupancy by Site
6
Site 1
Site 2
5
Number of Occupants
Site 3
Site 4
4
Site 5
Site 6
3
Site 7
Site 8
2
Site 9
Site 10
1
Site 11
Site 12
0
Site 13
Site 14
Month
Site 15
The houses ranged in size from 940 to 2,926 square feet (average of 1,936 sq ft), with between two to four bedrooms
and one to three bathrooms. Fourteen of the sites are serviced by public water supplies, while four sites have private
wells.
60
Table 22 summarizes the type and size of DHW system in use at each site as well as the water supply source and the
number of occupants. More details for each field test site can be found in Appendix A.
Table 22. Summary of 18 Field Test Sites
No.
Identifier
DHW System
Size
Water
Supply
Occupants
1
Ballina Rd, Cazenovia
GE HPWH with Recirculation Loop
Installed 4/10/2013
50 gal
Well
2 to 5
2
Gifford St, Syracuse
HE Condensing Gas Storage WH
Installed 5/1/2013
50 gal
Public
3
3
Brickyard Falls Rd,
Manlius
Tankless Gas Water Heater
Installed 3/18/2013
Tankless
Public
2
4
Burton St, Cazenovia
Tankless with Manifold and PEX and
Standard Gas Storage with copper
40 gal
Public
4 to 5
5
Hornady Dr, Syracuse
Gas Storage
40 gal
Public
2
6
Carrys Hill Rd,
Chittenango
Boiler with Indirect Tank
Well
4 to 5
7
Rathbun Rd,
Cazenovia
AO Smith Electric Storage
40 gal
Well
2 to 4
8
Astilbe Path, Liverpool
Rheem Condensing Tankless DHW and
Solar Hydronic Heating
120 gal
Public
1
9
Bunker Hill Way,
Syracuse (A)
Power Vent Tank
40 gal
Public
2
10
Bunker Hill Way,
Syracuse (B)
Power Vent Tank
40 gal
Public
2
11
Bunker Hill Way,
Syracuse (C)
Power Vent Tank
40 gal
Public
1 to 2
12
Burton St, Cazenovia
(B)
Standard Gas Storage
40 gal
Public
4
13
Tipp Hill, Syracuse
Standard Gas Storage
40 gal
Public
2
14
Fay Lane, Minoa
Power Vent Tank
40 gal
Public
2
15
Marcellus St, Syracuse
(A)
Power Vent Tank
40 gal
Public
2 to 3
16
Tioga St, Syracuse
Rinnai Gas Tankless
Tankless
Public
2
17
Marcellus St, Syracuse
(B)
Quiet Side Combi System HW and Infloor heating
Tankless
Public
3 to 4
18
Ridge Rd, Cazenovia
Boiler with Indirect Tank & small Elec.
Tankless
40 gal
Well
3
As discussed above in Section 3.2, there were three different types of datalogger systems installed at the 18 sites.
Details about the data loggers used at each site are given in Appendix A. A Campbell CR1000 datalogger was
installed at Sites 1 to 5. A Campbell CR200x datalogger was installed at eight sites to measure hot water use at 1561
minute intervals as well as at 5-second intervals during hot water draws. The other five sites used a Campbell CR10 or CR-10x datalogger to collect hot water flow data at 15-minute intervals.
5.2
Hot Water Usage
5.2.1 Hot Water Use Profiles
The total hot water use for each 15-minute interval was collected at all 18 sites. The plots below show the average
profiles at sites only during periods when the household had two occupants (this includes sites 1, 3, 5, 7, 9, 10, 11,
13, 14, 15 and 16). The daily profiles in Figure 46, Figure 47, and Figure 48 show hot water use profiles for a cold
month (February), a swing month (May), and a hot month (August), respectively. The average profile for each site
is plotted with a colored line while the overall average is shown with a thick black line. Data from days with no
recorded hot water use for the entire day (i.e., vacations) were excluded. Additional plots for individual sites are
available in Appendix C.
Figure 46. Hot Water Use Profile for 2-Person Home during a Cold Month (February 2013 & 2014)
15-Minute Daily Profile for 2-Person Home in February
0.50
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Flow (gpm)
0.40
11 Sites
Average Daily Total: 40.8 gal/day
Daily Range: 0.0 to 258.0 gal/day
1
3
5
7
9
10
11
13
14
15
16
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
Hour of Day
62
16
18
20
22
24
Figure 47. Hot Water Use Profile for 2-Person Home during a Swing Month (May 2013 & 2014)
15-Minute Daily Profile for 2-Person Home in May
0.50
Site
Site
Site
Site
Site
Site
Site
Site
Site
Flow (gpm)
0.40
9 Sites
Average Daily Total: 35.8 gal/day
Daily Range: 0.0 to 153.1 gal/day
1
3
5
9
10
11
13
14
15
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
16
18
20
22
24
Hour of Day
Figure 48. Hot Water Use Profile for 2-Person Home during a Hot Month (August 2013 & 2014)
15-Minute Daily Profile for 2-Person Home in August
0.50
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Flow (gpm)
0.40
10 Sites
Average Daily Total: 31.7 gal/day
Daily Range: 0.0 to 138.7 gal/day
1
3
7
9
10
11
13
14
15
16
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
Hour of Day
63
16
18
20
22
24
The plots below show the average profiles at sites when the household had four occupants for all or part of the time
(this includes sites 1, 4, 6, 7, 12, 17, and 18). The daily profiles in Figure 49, Figure 50, and Figure 51 show hot
water use profiles for a cold month (February), a swing month (May), and a hot month (August), respectively. The
average profile for each site is plotted with a colored line while the overall average is shown with a thick black line.
Data from days with no recorded hot water use for the entire day (i.e., vacations) were excluded. Additional plots
for individual sites are available in Appendix C.
Figure 49. Hot Water Use Profile for 4-Person Home during a Cold Month (February 2013 & 2014)
15-Minute Daily Profile for 4-Person Home in February
0.50
Site
Site
Site
Site
Site
5 Sites
Average Daily Total: 84.6 gal/day
Daily Range: 1.3 to 251.5 gal/day
4
7
12
17
18
Flow (gpm)
0.40
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
Hour of Day
64
16
18
20
22
24
Figure 50. Hot Water Use Profile for 4-Person Home during a Swing Month (May 2013 & 2014)
15-Minute Daily Profile for 4-Person Home in May
0.50
Site
Site
Site
Site
Site
5 Sites
Average Daily Total: 79.8 gal/day
Daily Range: 0.0 to 186.5 gal/day
1
4
7
17
18
Flow (gpm)
0.40
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
16
18
20
22
24
Hour of Day
Figure 51. Hot Water Use Profile for 4-Person Home during a Hot Month (August 2013 & 2014)
15-Minute Daily Profile for 4-Person Home in August
0.50
Site
Site
Site
Site
Site
5 Sites
Average Daily Total: 52.9 gal/day
Daily Range: 0.0 to 191.3 gal/day
1
6
7
17
18
Flow (gpm)
0.40
0.30
0.20
0.10
0.00
0
2
4
6
8
10
12
14
Hour of Day
65
16
18
20
22
24
5.2.2 Average Daily Hot Water Use
Using the 15-minute data we were able to calculate the average HW use for each site for individual days when the
home was occupied. Table 23 shows the beginning and ending dates of data collection for all 18 field sites, along
with the average, median, and standard deviation of HW use (gal/day) at each site.
Table 23. Beginning and Ending Dates for Data Collection at Sites with 15-minute Monitoring
Site
Occupancy
Data Start Date
Data End Date
Average
HW Use
(gal/day)
Median
HW Use
(gal/day)
St Dev
HW Use
(gal/day)
Site 1
Variable (1-5)
12/1/2012
12/31/2014
50.2
45.5
34.9
Site 2
3
12/14/2012
5/25/2014
75.8
67.0
48.8
Site 3
2
12/13/2012
12/31/2014
30.8
28.4
18.1
Site 4
Variable (4-5)
12/10/2012
7/18/2014
124.0
125.5
38.4
Site 5
2
12/14/2012
12/31/2014
55.6
54.6
22.9
Site 6
Variable (4-5)
1/24/2013
1/3/2014
24.9
23.8
14.1
Site 7
Variable (2-4)
1/24/2013
2/21/2014
42.5
40.7
23.3
Site 8
1
12/1/2012
2/4/2013
48.2
47.1
17.5
Site 9
2
1/19/2013
2/3/2014
21.7
20.6
10.8
Site 10
2
12/1/2012
2/19/2014
46.7
44.9
17.7
Site 11
Variable (1-2)
1/19/2013
11/27/2014
27.8
27.1
13.7
Site 12
4
12/1/2012
4/5/2013
77.3
76.9
24.3
Site 13
2
1/10/2013
2/20/2014
54.1
52.0
16.2
Site 14
2
12/1/2012
2/4/2014
56.5
56.4
17.7
Site 15
Variable (2-3)
12/14/2012
12/4/2014
25.9
23.4
18.9
Site 16
2
6/18/2013
2/19/2014
12.6
10.4
7.9
Site 17
Variable (3-4)
1/24/2013
2/25/2014
81.0
74.3
41.1
Site 18
Variable (3-4)
12/14/2012
2/21/2014
49.2
50.3
18.4
2-person
2
12/1/2012
12/31/2014
39.9
38.9
23.8
4-person
4
12/1/2012
5/10/2014
73.8
66.2
45.9
All Sites
Variable (1-5)
12/1/2012
12/31/2014
51.7
45.2
37.5
Table 23 above calculates average hot water use excluding days where there was no hot water use (i.e., vacation or
vacant days). Table 24 determines the average hot water use including these vacant or no-water-use days. The
overall average hot water use decreases by 0.5 gallons per day when vacant days are included. The sites with more
vacant days saw larger changes in the average daily use, as would be expected. All other plots and graphs in this
report using 15-minute data are based on averages that exclude these vacant days and correspond to Table 23.
66
Table 24. Impact of Including Vacation and Missing Days in Hot Water Use Calculation
Site
Occupancy
Occupied
Days
Vacant
Days
Excluding Vacant Days:
Average HW Use (gal/day)
Including Vacant Days:
Average HW Use (gal/day)
Site 1
Variable (1-5)
697
33
50.2
47.9
Site 2
3
528
2
75.8
75.5
Site 3
2
693
25
30.8
29.7
Site 4
Variable (4-5)
569
1
124.0
123.8
Site 5
2
709
9
55.6
54.9
Site 6
Variable (4-5)
315
1
24.9
24.9
Site 7
Variable (2-4)
394
1
42.5
42.5
Site 8
1
66
1
48.2
48.2
Site 9
2
381
1
21.7
21.7
Site 10
2
439
7
46.7
46.0
Site 11
Variable (1-2)
568
10
27.8
27.3
Site 12
4
119
1
77.3
77.3
Site 13
2
402
6
54.1
53.3
Site 14
2
424
7
56.5
55.5
Site 15
Variable (2-3)
695
5
25.9
25.7
Site 16
2
245
2
12.6
12.5
Site 17
Variable (3-4)
398
1
81.0
80.8
Site 18
Variable (3-4)
421
14
49.2
47.7
67
Figure 52 uses the 15-minute data to show the median hot water daily use as a symbol with plus-or-minus one
standard deviation about the average shown by the vertical line. The aggregate 2-person house had an average,
median, and standard deviation of 39.9, 38.9, and 23.8 gallons per day, respectively. For the 4-person data set, the
average, median, and standard deviation were 73.8, 66.2, and 45.9 gallons per day, respectively.
Figure 52. Median Hot Water Use at Each Site with a Range of Plus or Minus One Standard
Deviation about the Mean
Median Daily Volume (+/- Std Dev) By Site
150
2-Person Data
4-Person Data
Mixed/Other Data
HW Use (gal/day)
100
Average 4-Person
50
Average 2-Person
0
1
2
3
4
5
6
7
8
9
10
11
Site Number
68
12
13
14
15
16
17
18
Daily average hot water use for all households with two occupants and for households with four occupants is shown
in Figure 53. The average value was calculated independently based on each valid day of data for each site with the
correct number of occupants on a given day during each month. Days with zero hot water use were excluded from
the calculations.
Figure 53 clearly shows that hot water use varies seasonally for both the 2-person and 4-person households. This
trend is primarily due to occupants (or machines) mixing in more hot water when the incoming ground or city water
is colder in the winter. The amount of hot water required to reach 100°F temperature at the shower head is greater
when cold water is at a lower temperature in the winter. Seasonal cold water trends are shown later in this
document.
Figure 53. Average Daily Hot Water Use for 2-Person and 4-Person Homes
Average Daily Hot Water Use
250
2-Person Households (11 Sites)
4-Person Households (7 Sites)
Hot Water Use (Gal/Day)
200
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
69
Aug
Sep
Oct
Nov
Dec
Figure 54 compares the average water use for each two-person household to the overall average. The plot confirms
there is significant variation between two-person households.
Figure 54. Average Daily Hot Water Use for 2-Person Households Compared to the Average
Average Daily Hot Water Use for 2-Person Household (11 Sites)
250
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Hot Water Use (Gal/Day)
200
1 Av erage
3 Av erage
5 Av erage
7 Av erage
9 Av erage
10 Av erage
11 Av erage
13 Av erage
14 Av erage
15 Av erage
16 Av erage
Overall Average
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Besides household-to-household variation, occupants in a given house use different amounts of hot water on a dayto-day basis. This can be explained by the different activities that occupants engage in on different days. Figure 55
shows the data for sites with two occupants. For each month the overall average, plus-or-minus one standard
deviation, is shown as a blue line for the month. The median, maximum, and minimum are shown in black. The
data for each month on the plot only includes days where hot water use was greater than zero and the number of
occupants was equal to two. The average number of sites included in each monthly calculation is shown at the top
of the plot. For some months, days were excluded because they did not have two occupants or because usage was
zero. Each monthly value is based on data from 5 to 7 sites (out of a possible 11 sites).
The variation in average daily hot water use between sites is again shown in Figure 56 in a slightly different way. In
this case the average for each site for each month is shown as a colored symbol and a colored vertical line represents
the minimum and maximum daily hot water use in the month for that site. The black trend line super imposes the
overall average for all the two-person sites.
70
Figure 55. Average Daily Hot Water Use for 2-Person Households
Average Daily Hot Water Use for 2-Person Household (11 Sites)
250
7.1
6.6
5.9
5.8
5.0
5.1
5.3
5.5
6.6
6.3
5.7
4.8
Nov
Dec
Median
Average +/- 1 Std. Dev.
200
Hot Water Use (Gal/Day)
Maximum / Minimum
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Figure 56. Average Daily Hot Water Use for 2-Person Households with Each Site Shown
Average Daily Hot Water Use for 2-Person Household (11 Sites)
250
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Site
Hot Water Use (Gal/Day)
200
1 Av erage
3 Av erage
5 Av erage
7 Av erage
9 Av erage
10 Av erage
11 Av erage
13 Av erage
14 Av erage
15 Av erage
16 Av erage
Overall Average
Site Average with Min / Max Bar
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
71
Sep
Oct
Nov
Dec
Figure 57 compares the average water use for each four-person household to the overall average. The plot confirms
there is significant variation between the four-person households.
As discussed above, household-to-household variation in four-person homes is also significant. Occupants in a
given house use different amounts of hot water on a day-to-day basis, since they engage in different activities on
different days. Figure 58 shows the data for sites with four occupants. For each month the overall average plus-orminus one standard deviation is shown in blue. The median, maximum, and minimum are shown in black. The data
for each month on the plot only includes days where the total hot water use was greater than zero and the number of
occupants was equal to four. The average number of sites included in each monthly calculation is shown at the top
of the plot. For some months, days were excluded because they did not have four occupants or usage was zero.
Each monthly value is based on data from 1 to 3 sites. Therefore, the results in the four-occupant households may
have a higher uncertainty than for two person households.
The variation in average daily hot water use between sites is again shown in Figure 59. In this case the average for
each site for each month is shown as a colored symbol and a colored vertical line represents the minimum and
maximum daily hot water use in the month for that site. The thick black line super imposes the overall average for
all the four person sites.
Figure 57. Average Daily Hot Water Use for 4-Person Households Compared to the Average
Average Daily Hot Water Use for 4-Person Household (7 Sites)
250
Site
Site
Site
Site
Site
Site
Site
Hot Water Use (Gal/Day)
200
1 Av erage
4 Av erage
6 Av erage
7 Av erage
12 Av erage
17 Av erage
18 Av erage
Overall Average
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
72
Sep
Oct
Nov
Dec
Figure 58. Average Daily Hot Water Use for 4-Person Households
Average Daily Hot Water Use for 4-Person Household (7 Sites)
250
2.5
2.8
2.8
2.3
1.7
1.4
1.2
1.7
0.9
1.0
1.0
1.6
Oct
Nov
Dec
Median
Average +/- 1 Std. Dev.
200
Hot Water Use (Gal/Day)
Maximum / Minimum
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Figure 59. Average Daily Hot Water Use for 4-Person Household with Each Site Shown
Average Daily Hot Water Use for 4-Person Household (7 Sites)
250
Site
Site
Site
Site
Site
Site
Site
Hot Water Use (Gal/Day)
200
1 Av erage
4 Av erage
6 Av erage
7 Av erage
12 Av erage
17 Av erage
18 Av erage
Overall Average
Site Average with Min / Max Bar
150
100
50
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
73
Sep
Oct
Nov
Dec
5.3
Entering Cold Water Temperatures
As mentioned above, the entering cold water temperatures from a well or public water supply vary across the year.
Cold water temperatures were measured at six sites (sites 1, 2, 3, 4, 5, and 8). The entering water temperature for
each site could be inferred two different ways:


using a flow-weighted average calculation in the data logger at the scan interval, or
using the daily minimum entering water temperature from the 15-minute data
Figure 60 shows a comparison between the two values. In theory the flow weighted values should be more correct,
yet delays in the temperature sensor response could sometimes lead to skewed values. We judged that the minimum
value observed in a day – which typically corresponded to a large hot water draw – provides the best indication of
the true entering water temperature for that day.
Figure 60. Daily Minimum Entering Water Temperature vs. Daily Flow-Weighted Average EWT
Entering Water Temperature
Daily Flow-Weighted Average EWT (F)
80
70
60
Site 1 (Private)
50
Site 2 (Public)
Site 3 (Public)
Site 4 (Public)
Site 5 (Public)
40
Site 8 (Public)
30
30
40
50
60
70
80
Daily Minimum EWT (F)
Figure 61 shows the daily entering water temperatures for the six sites where it was measured. The sites on a public
or city water supply show a much larger seasonal variation in entering water temperature than for the home with a
private well. The seasonal fluctuation of the city water supply is 25-30°F across the year, since these pipes are often
closer to the surface. Site 1, which has a private supply from a well that is more than 200 ft deep, showed much less
74
variation across the year. Figure 62 shows the relationship between daily water entering temperature (minimum
EWT) and daily hot water use. Because of the day-to-day variations at each site it is difficult to discern a trend.
However, overall most of the sites might imply lower hot water use with warmer inlet temperatures.
Figure 61. Seasonal Variation in Entering Water Temperature
Entering Water Temperature
80
Site 1 (Private)
Site 2 (Public)
Site 3 (Public)
Site 4 (Public)
Site 5 (Public)
Site 8 (Public)
Daily Minimum EWT (F)
70
60
50
40
30
Q4 Q1
2013
Q2
Q3
Q4
Q1
Q2
2014
75
Q3
Q4
2015
Figure 62. Daily Hot Water Use vs. Entering Water Temperature
Daily Hot Water Use
400
Site 1 (Private)
Site 2 (Public)
Site 3 (Public)
Site 4 (Public)
Daily Hot Water Use (gallons)
Site 5 (Public)
Site 8 (Public)
200
0
30
40
50
60
70
80
Daily Minimum EWT (F)
5.4
Hot Water Draw Events
At the thirteen (13) sites listed in the table below, data were collected at 5-second intervals so that each water draw
event could be separately identified. Table 25 also summarizes the average hot water use as well as the number of
draw events per day through June 2014. Some minor differences between the average use given in this table and in
the 15-minute data table (Table 23) are noted. Appendix C provides histograms and other statistics of the event data
for each site and each household size (2-5 persons) for all the 5-second data sites.
76
Table 25. Beginning and End Dates for Event Data at Each Site with 5-second Monitoring
Site
Occupancy
Data
Data
Average
Average
Start Date
End Date
HW Use
(gal/day)
Draw
Events
83
(No/day)
Site 1
Variable (2-5)
12/01/2012
02/13/2014
53a
Site 2
3
12/14/2012
05/25/2014
77
52
Site 3
2
12/13/2012
12/31/2014
31
24
Site 4
Variable (4-5)
12/10/2012
07/04/2014
122b
280
Site 5
2
12/14/2012
12/31/2014
57
39
Site 10
2
12/01/2012
02/19/2014
47
48
244
Site 12
4
12/20/2012
04/04/2013
63c
Site 13
2
01/10/2013
02/16/2014
54
44
Site 14
2
12/01/2012
02/04/2014
57
34
76
Site 15
Variable (2-3)
12/14/2012
12/05/2014
27d
Site 16
2
01/18/2013
02/19/2014
12
35
Site 17
Variable (3-4)
01/24/2013
02/25/2014
80
65
Site 18
Variable (3-4)
12/14/2012
02/21/2014
49
91
Notes:
a- Site 1’s 15-minute average HW use differed from the 5-second average because 5-second data was not collected
through June 2014.
b- Site 4’s 15-minute average HW use differed from the 5-second average because 5-second data was not collected
on 1/09/2013, 1/30/2013, and 2/06/2013.
c- Site 12’s 15-minute average HW use differed from the 5-second average because 5-second data was not collected
from 1/12/13 to 2/13/2013.
d- Site 15’s 15-minute average HW use differed from the 5-second average because 5-second data was not collected
from 2/14/2013 to 4/01/2013.
Figure 63 shows the data for all 5-second interval sites, plotting the daily hot water use versus the number of hot
water draw events. Each site is shown as a different color. Sites with 2 people are shown as “+” while sites with 4
people are shown as a “□”. There is significant scatter in the data, though some sites do show similar patterns on the
plot.
Figure 64 reduces each site to a single data point, using the median hot water use and median number of draws (or
events). In this case, sites with 2 people are shown as “Δ” while sites with 4 people are shown as a “□”. Sites with
variable occupancy are shown with a “*”. Each site is identified by its own color. The large symbols indicate the
medians for the 2-person and 4-person households. The median 2-person household uses 43 gallons per day with 37
draw events. The median 4-person household uses 80 gallons per day with 120 draw events.
77
Figure 63. Hot Water Use vs. Number of Draw Events (all days through December 2014)
Daily Hot Water Use (5423 days plotted)
250
2-Person Home
4-Person Home
Other
HW Use (gal/day)
200
150
100
50
0
0
100
200
300
400
500
Number of Events
Site 1
Site 2
Site 3
Site 4
Site 5
Site 10
Site 12
Site 13
Site 14
Site 15
Site 16
Site 17
Site 18
Figure 64. Hot Water Use vs. Number of Draw Events (Median for Sites through December 2014)
Median Daily Hot Water Use (13 sites plotted)
150
2-Person Data
4-Person Data
Mixed Data
HW Use (gal/day)
100
50
All 4-Person Data
All 2-Person Data
0
0
100
200
300
Number of Events
Site 1
Site 2
Site 3
Site 4
Site 5
Site 10
Site 12
Site 13
78
Site 14
Site 15
Site 16
Site 17
Site 18
.
5.5
Identifying Draws and Estimating Hot Water Waste
Energy is wasted every time there is a hot water draw. Hot water runs through distribution piping to the fixture for
the occupant’s use. After the draw is complete, the hot water in that pipe cools down to ambient conditions over
several hours. The next time hot water is required at that fixture the occupant runs the previously-heated water
down the drain while waiting for hot water to appear at the fixture again. Because this process is repeated at several
fixtures throughout each day, hot water waste is estimated to be on the order of 10%–30% of total hot water use.
We installed temperature sensors near each fixture on the distribution systems at the Field Test Sites 1 through 5 as
described in Section 3.2 and Figure 5. The temperature sensor near each fixture was used to determine the portion
of hot water draw that was deemed useful, using the methods described in Section 3.2.2. This analysis is fully
described summarized in a separate Building America Report (Henderson and Wade, 2014). A sample of the results
from that report is summarized here.
Figure 63 shows one flow event with temperatures corresponding to the master bath shower. The plot in the upper
left shows the flow trace during the event at 5 second intervals (ending with five intervals at zero flow). The red line
corresponds to the most common flow rate (in gpm) or “mode” during the event. The plot in the lower left shows the
various temperatures during the event. Blue lines correspond to trunk temperatures (TT1, TT2…) while the green
lines are the fixture temperatures (TF1, TF2,..). Each line is identified by a number. The plot at the upper right
shows the temperature rise for each sensor compared to the beginning of the event (using the same colors). The red
line on the upper right and lower left plots is the supply temperature from the water heater unit (TS). The black line
on these plots corresponds to the cold water inlet (TC). The trunk line with the highest temperature rise is indicated
by blue text and the fixture temperature with the biggest rise is indicated with green text.
This first shower of the day lasts for about 6 minutes (84 intervals) and the total draw is 8.8 gallons. The temperature
leaving the tank increases as expected and the inlet cold water temperature drops as cold water flows by the sensor.
The trunk to the master bedroom increases by 40°F as expected. In this case, there is no temperature sensor for the
shower fixture, only for the bathroom sink. So a trunk temperature rise with no large change in fixture temperature
indicates a shower. The automated procedure in this case classified this event as a draw at the “Master Shower”
fixture. Of the 8.79 gallons of hot water used, only 8.03 gallons was deemed as useful by the criteria (in this case
the trunk temperature reaching 100°F).
79
Figure 65. A Sample Hot Water Draw is Identified and Evaluated at Site 1
12/02/12 08:00:05 GAL = 8.79
TH - Hot
Temperature Rise (F)
Flow Rate (gpm)
2.0
1.5
1.0
0.5
0.0
TC = Cold
30
20
10
2
6
57
4
3
5
4
3
2
1
8
0
-10
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
:0
:1
:2
:3
:4
:5
:6
:7
:0
:1
:2
:3
:4
:5
:6
:7
12/02/12 08:00:05 Intervals=84
140
Temperature (F)
1
120
TT1: Trunk Temp - Master Bathroom: Rise= 40.7
TT2: Trunk Temp - Bathroom 2: Rise= 5.8
1
100
Master Shower
80
26
3
4
3
8
5
1
57
2
4
60
40
0.
0.
0.
0.
0.
0.
0.
0.
:0
:1
:2
:3
:4
:5
:6
:7
Useful GAL = 8.03 (TT > 100. )
TF6: Fixture Temp 6 - Master Bath Sink: Rise= 3.8
TF7: Fixture Temp 7 - Bathroom 2 Sink: Rise= 0.3
DHW Site 1
The results from repeating this analysis for all 8,534 events that occurred over 123 days at Site 1 are shown in Table
26. This home has two adults and two college-age students who returned for the college breaks (who mostly used
bathroom 2). Fewer than half of the 8,534 events in the 123-day period could be assigned. However, these classified
(or assigned) events accounted for 94% of the total water use. Numerous small events could not be assigned. The
average unclassified event was less than 0.1 gal. Of the unclassified events, only 18 were over more than 1 gal in the
123-day period. Overall, about 82% of the total classified water use was deemed to be useful, based on the pipe
temperature threshold. As expected, the sinks result in the largest amount of hot water waste on a percentage basis.
80
Table 26. Summary of Hot Water Draw Identification at Site 1
December 1, 2012 to April 2, 2013 (123 Days, 8,354 events)
Events per
Day
Total Hot
Water Use
(gpd)
Useful Hot
Water (gpd)
Wasted Hot
Water (gpd)
Useful %
Dishwasher
1.5
2.7
2.4
0.3
90%
Kitchen Sink
17.0
9.1
5.7
3.3
63%
Half Bath Sink
2.1
0.7
0.1
0.5
21%
Washing Mach
0.7
9.1
8.3
0.8
91%
Utility Sink
0.3
0.9
0.5
0.4
53%
Master Sink
6.8
2.2
0.6
1.5
28%
Master Shower
1.5
17.8
17.0
0.8
95%
Bath 2 Sink
1.7
1.2
0.6
0.6
49%
Bath 2 Shower
0.7
8.4
7.7
0.8
91%
Unaccounted
36.8
3.4
–
3.4
0%
All Events
69.4
55.4
42.9
12.5
77%
Classified
33
52.0
42.9
9.1
82%
% Classified
47%
94%
100%
73%
This process was repeated at all five field test sites. Overall, the process was able to classify less than half of the
water draws in the five homes but the classified draws accounted for about 95% of the hot water volume in each
home. This classification process was achieved without the need to install flow meters for each fixture (which
would have been cost-prohibitive and more disruptive to homeowners). The average number of events at each home
ranged from 26 per day to 180 per day.
A temperature of 90°F at the fixture piping (or 100°F at the trunk line) was selected as the threshold for gauging
usefulness. The amount of hot water deemed to be useful ranged from a low of 75% at Site 4 to a high of 91% at
Site 5—thus implying 9%–25% waste. Site 4 may have had a lower useful percentage in part be due to the fact that
the sensors installed on cross-linked polyethylene (PEX) piping may have had a slower thermal response. As
expected, the amount of hot water waste was found to be higher for bathroom and kitchen sinks and lowest for
showers and washing machine draws.
81
5.6
Field Test Site 1 – HPWH Performance
Field Site 1 had a General Electric 50 gallon Heat Pump Water Heater (model GEH50DEEDSR) installed on April
10, 2013. Figure 66 shows that the daily conversion efficiency (thermal energy out divided by electric energy in) for
the unit as function of hot water use. Occasionally a large water draw resulted in operation of the electric elements.
These days are shown with a red diamond. The efficiency is significantly reduced on these days. Days with
element operation are not strongly correlated to the daily hot water use. The blue *’s indicate days when the
compressor was off and the unit operated like a conventional electric water heater. The green diamonds indicate
very small amounts of element operation which are discussed in more detail below.
Figure 67 shows the same data but with input power as function of water use. The power shows a strong linear trend
with hot water use – except on days with element operation. The linear trend of electric use projects back to 2
kWh/day with no hot water use. This compares to the value of 0.69 kWh/day measured in the laboratory. The
operation of the recirculation pump may account for most of the additional standby losses.
Table 27 summarizes the monthly performance of the HPWH. The average annual conversion efficiency for May
2013 to April 2014 was 1.4 with an average water use of 43.7 gallons per day. The nominal or rated EF for this unit
is 2.4. The monthly data shows that the HPWH compressor stopped working in August 2014 and reverted to
resistance element operation.
Resistance-only operation provided the opportunity to directly compare hybrid heat pump operation and resistanceonly operation. Comparing the month efficiencies for December 2013 and 2014, which had similar water use,
shows the conversion efficiencies were 1.37 and 0.67 respectively. This comparison implies that the HPWH reduces
electric use compared to standard electric operation by one half.
82
Figure 66. Conversion Efficiency (or COP) as a Function of Hot Water Use
DHW1 - 04/10/13 to 03/31/15
3.0
Conversion Efficiency (-)
2.5
2.0
Elements ON
1.5
1.0
0.5
0.0
0
50
100
150
200
Hot Water Use (gal/day)
Figure 67. HPWH Input Power as a Function of Hot Water Use
DHW1 - 04/10/13 to 03/31/15
30
Input Power (kWh/day)
25
Elements ON
20
15
10
5
0
0
50
100
150
Hot Water Use (gal/day)
83
200
Table 27. Summary of Monthly HPWH Performance
HW Use
(gal/day)
Electric
Use
(kWh)
COP (-)
Element
Runtime
(hrs)
Vacation
Days
May-13
69.3
182.9
1.71
9.6
0
Jun-13
60.5
139.4
1.80
5.6
2
Jul-13
39.2
108.2
1.51
3.6
7
Aug-13
44.7
128.9
1.47
5.7
9
Sep-13
37.3
101.7
1.49
2.7
2
Oct-13
25.6
97.9
1.10
3.2
9
Nov-13
45.4
133.5
1.42
6.9
0
Dec-13
61.1
200.4
1.37
19.1
1
Jan-14
29.9
120.5
1.12
6.1
10
Feb-14
34.9
117.5
1.22
4.7
8
Mar-14
45.0
155.6
1.35
7.4
1
Apr-14
31.3
125.5
1.10
7.2
9
May-14
61.6
167.0
1.65
7.5
0
Jun-14
32.3
112.7
1.22
4.7
6
Jul-14
68.1
177.2
1.65
11.4
4
Aug-14
62.2
155.4
1.71
5.8
0
Sep-14
43.7
277.2
0.66
56.0
4
Oct-14
39.1
294.5
0.57
65.7
3
Nov-14
38.2
290.0
0.57
65.1
6
Dec-14
59.5
418.3
0.67
94.2
0
Jan-15
22.7
267.9
0.40
60.1
7
Feb-15
41.8
319.4
0.57
71.9
2
Mar-15
29.6
306.2
0.47
68.4
10
Annual
43.7
1,612.0
1.41
81.7
58
Several days with electric element operation are shown in the following figures. Each shows the hot water draw
pattern, the operating status of the upper (SE1) and lower (SE2) elements, and power use profile for each day. It
typically takes a large draw to induce element operation. The numbers displayed on the water use plot indicate the
number of gallons associated with each draw event.
Figure 68 shows that on April 10, 2013 a 20 gallon draw following a 9.8 gallon draw was sufficiently large to
activate the element. Once the elements start, they operate for 80-90 minutes (at 4.5 kW). Figure 69 shows another
large draw (or series of draws) on April 23 that also induced element operation. Figure 70 shows operation for May
21 where large draws within a few hours of each other induced two distinct cycles of element operation.
84
Figure 71 shows a day with only a few minutes of element operation that occurred for no apparent reason. This
unexpected behavior started to occur more frequently in the fall of 2013.
Figure 68. Plot of HPWH Performance for a Day with Element Operation - April 10, 2013
04/10/13
FW= 41.4 gal/day
12
9.8
HW Use (gal)
10
DRAW= 20.0 gal
0.9
4.0
20.0
1.2
2.9
1.3
0.9
0.4 0.0
8
6
4
2
0
22:
0:
9
10
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
11
SE1= 25.8 min, SE2= 56.9 min
ON
SE2
OFF
ON
SE1
OFF
22:
0:
9
10
WE=
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
11
04/10/13
8.0 kwh/day
5
Power (kW)
4
3
2
1
0
22:
0:
9
10
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
11
April
2013
85
Figure 69. Plot of HPWH Performance for a Day with Element Operation - April 23, 2013
04/23/13
FW= 101.2 gal/day
HW Use (gal)
25
1.2
0.1
7.6
20
22.3
DRAW= 54.2 gal
0.0
1.1
11.9
0.7
54.2
0.2
1.6
0.3
15
10
5
0
22:
0:
22
23
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
24
SE1= 18.4 min, SE2= 63.3 min
ON
SE2
OFF
ON
SE1
OFF
22:
0:
22
23
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
24
04/23/13
WE= 10.7 kwh/day
5
Power (kW)
4
3
2
1
0
22:
0:
22
23
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
24
April
2013
86
Figure 70. Plot of HPWH Performance for a Day with Significant Element Operation - May 21, 2013
05/21/13
FW= 114.6 gal/day
HW Use (gal)
15
0.1
10.1 0.1
1.1
DRAW= 64.7 gal
1.0
1.1
64.7
0.2
33.7
0.0
0.0
0.2
2.3
10
5
0
22:
0:
20
21
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
22
SE1= 14.2 min, SE2= 130.5 min
ON
SE2
OFF
ON
SE1
OFF
22:
0:
20
21
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
22
05/21/13
WE= 14.1 kwh/day
5
Power (kW)
4
3
2
1
0
22:
0:
20
21
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
22
May
2013
87
Figure 71. Plot of HPWH Performance for a Day with Small Amounts of Element Operation November 12, 2013
11/12/13
FW= 34.1 gal/day
HW Use (gal)
10
0.2
0.2
6.0
8
DRAW= 17.7 gal
17.7
0.3
4.7
2.9
0.2
0.30.1
6
4
2
0
22:
0:
11
12
SE1=
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
0.0 min, SE2=
2.8 min
ON
SE2
OFF
ON
SE1
OFF
22:
0:
11
12
WE=
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
11/12/13
3.1 kwh/day
Power (kW)
0.8
0.6
0.4
0.2
0.0
22:
0:
11
12
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
13
November
2013
88
All days through May 28, 2014 with any element operation are shown on Figure 72. The electric elements are
usually activated by a 30 gallon draw. In some cases draws as small as 20 gallons activated the elements – typically
when a smaller draw had just recently occurred (and the total of the two draws was near to or exceeding 30 gallons).
Figure 72. Plot of Element Runtime as Function of Hot Water Draw Size
200
2 cycles of element
operation
Element Runtime (min)
150
100
1 cycle of element
operation
50
few minutes of unexplained operation
0
0
20
40
60
80
Largest Draw Volume (gal)
89
100
120
5.6.1 Unexpected Aspects of HPWH Performance
Figure 73 is a shade plot showing when the lower element (SE2) on the HPWH tank operated from October 2013
through February 2014. The periods with normal periods of element operation appear as the black 1-2 hour events.
The many small 2-6 minute periods of operation appear as light gray intervals. The small events started in late
October 2013 and continued through until late January 2014. Table 28 lists all the unexplained events that occurred
over the entire monitoring period. In addition to the periods on the shade plot, other grouping of events also
occurred in April 2014. The table also shows the water use for each of these days. The events do not appear to have
been linked to changes in daily water use. A portion of these events (shown in red) showed a different operating
pattern with 12-16 minutes of operation for the top element (SE1). These short periods of operation are also
unexpected relative to the expected control scheme of the unit.
Figure 73. Shade Plot Showing Unexplained Periods When Lower Electric Element Runs a Small
Amount
SE2
24
22
20
18
Hour of Day
16
14
12
10
8
6
Period with Short
Element Operation
4
2
0
October
November
December
2013
January
2014
Day (MAX/MIN =
90
15.00/ 0.00 )
Feb
Table 28. Summary of Days with Unexpected Element Operation
Date
7/19/2013
10/29/2013
11/3/2013
11/5/2013
11/10/2013
11/12/2013
11/13/2013
11/14/2013
11/19/2013
11/23/2013
11/24/2013
11/25/2013
11/26/2013
11/29/2013
11/30/2013
12/2/2013
12/3/2013
12/4/2013
12/7/2013
12/8/2013
12/9/2013
12/10/2013
12/11/2013
12/13/2013
12/14/2013
12/15/2013
12/16/2013
12/17/2013
12/18/2013
12/20/2013
12/21/2013
12/28/2013
12/30/2013
HW Use
(gal)
60.6
32.4
30.0
87.0
33.6
34.1
60.6
56.7
25.4
35.9
26.5
16.9
32.7
34.9
59.2
44.7
39.6
33.9
53.9
15.7
38.0
34.7
49.9
33.1
35.9
40.1
50.3
34.3
59.1
49.2
89.0
17.5
64.5
SE1
(min)
SE2
(min)
16.3
-
3.0
1.4
3.6
1.5
2.8
4.7
4.1
3.3
4.7
4.9
3.2
5.2
1.7
5.2
1.5
5.4
3.8
1.7
5.0
3.3
3.1
3.7
1.7
4.8
6.6
3.3
7.6
3.5
5.4
3.2
1.4
4.7
Date
1/2/2014
1/3/2014
1/4/2014
1/6/2014
1/7/2014
1/8/2014
1/9/2014
1/10/2014
1/12/2014
1/14/2014
1/15/2014
1/16/2014
1/18/2014
1/20/2014
2/9/2014
4/10/2014
4/11/2014
4/12/2014
4/13/2014
4/16/2014
4/18/2014
4/20/2014
4/21/2014
4/23/2014
4/24/2014
4/25/2014
4/26/2014
4/27/2014
4/28/2014
4/29/2014
4/30/2014
5/2/2014
5/3/2014
5/4/2014
5/5/2014
5/6/2014
5/7/2014
5/8/2014
5/11/2014
5/12/2014
5/20/2014
7/21/2014
HW Use
(gal)
24.5
0.0
1.6
0.0
0.0
1.8
0.1
0.0
1.4
0.0
0.2
4.9
84.0
77.7
52.1
53.9
8.0
58.3
113.3
29.2
26.2
41.4
32.9
32.0
40.6
44.0
59.8
26.6
10.9
33.9
71.1
21.3
23.1
45.0
44.2
69.1
45.5
49.0
101.2
SE1
(min)
SE2
(min)
12.3
12.8
12.7
13.8
3.6
1.4
1.5
1.7
3.3
1.6
1.8
1.8
5.8
1.7
3.5
5.0
3.4
3.4
1.7
1.6
6.3
5.1
3.5
5.0
5.4
6.6
6.8
6.3
5.2
6.8
6.5
4.8
1.4
1.6
6.8
3.9
3.2
2.9
1.8
1.7
2.1
3.1
-
Notes: Black data are short periods of lower element operation (SE2). Red data are days with upper element
operation (SE1)
91
Around August 3, 2014 the power draw of the HPWH compressor suddenly started to change. Figure 74 shows how
the compressor power started to fluctuate for a few days right before stopping operation on September 6. The
fluctuation in power might imply a slow leak of refrigerant charge or some other slowly-failing components such as
the evaporator fan. From that point on the resistance elements satisfied the water heating load.
Figure 74. Last Few Days of HPWH Compressor Operation
5
Element Power
(4.5 kW)
Power (kW)
4
3
2
Normal
Compressor
Power
1
0
1
2
3
Fluctuating
Compressor
Power
4
5
6
7
8
9
10
11
12
September
2014
5.6.2 HPWH Interactions with the Space
The psychrometric plot in Figure 75 below shows two distinct groupings of the daily data:


One trend corresponding to the summer period when humidity is high.
One trend in the winter when the basement dries out (and becomes colder) because outdoor conditions are
dry and cold.
Figure 76 shows the daily wet bulb and dry bulb in basement. While the design conditions for EF test specify 68°F
and 50% RH (which corresponds to a wet bulb near 57°F), the conditions in this basement were often much colder.
92
Figure 75. Psychrometric Conditions in Basement – with and without HPWH Operation
DHW1 - 11/16/12 to 05/28/14
0.020
HPWH ON
0.015
Humidity Ratio (lb/lb)
HPWH OFF
Dehumidifier
set at 60% RH
0.010
80%
DOE Test Conditions
(68°F, 50%)
60%
0.005
40%
Winter Period
(2012/2013, before HPWH)
20%
50
55
60
65
70
75
0.000
80
Dry Bulb Temperature (F)
Figure 76. Wet Bulb and Dry Bulb in Basement – with and without HPWH Operation
DHW1 - 11/16/12 to 05/28/14
65
HPWH ON
60
Wet Bulb Temperature (F)
HPWH OFF
55
DOE Test
Conditions
(68°F, 50%)
50
45
Winter Period
(2012/2013, before HPWH)
40
55
60
65
Basement Temperature (F)
93
70
Figure 77 shows effect of HPWH and dehumidifier operation on temperature and humidity conditions in the
basement. There were minimal temperature and humidity variations before the HPWH was installed on April 10,
2013. Once the HPWH started to operate the temperature near the units started to vary by as much 10°F. As the
temperature decreased the dew point decreased as well, but the relative humidity increases as the temperature is
depressed. Dehumidifier (DH) operation started in late May 2013. However the dehumidifier runtime sensor was
not installed until Jun 12, 2013. DH runtime is shown on the bottom of the figure.
Figure 77. Basement Temperature and Humidity Shown with DH Operation – Entire Test Period
Basement Temperature Trends
60
40
Crawl Space
Near Unit
Dew Point (Near Unit)
Temperature
Run Time (min)
20
Q4
Q1
2012
2013
Q2
Q3
Q4
Q1
Q2
2014
Dehumidifier Status
Basement Temperatures and RH
15
80
100
10
80
60
Dehumidifier Set Point: 60% RH
5
60
40
0
20 Q4
2012
Q4
Q1
Q2
2013
Q1
Q2
2012
2013
Basement Temp (Near Unit)
Basement RH (Near Unit)
Q3
Q4
Q3
Q4
40
Q1
Q2
2014
Q1
Q2
2014
Dehumidifier Status
Run Time (min)
15
10
5
0
Q4
Q1
2012
2013
Q2
Q3
Q4
Q1
2014
94
Q2
20
Relative Humidity (%)
Temperature (F)
80
Figure 78 shows the same data for one day on June 20. Each time the HPWH runs the temperature near the unit
drops by 5-10°F. The dew point also drops. However, the relative humidity increases as the space is cooled. This
increase in relative humidity causes the dehumidifier to stay on continuously for the 3 hour period when the HPWH
is on (8-11 am). In the short term, HPWH operation induces additional DH operation. Longer terms impacts are
discussed below.
Figure 78. Basement Temperature and Humidity Shown with DH Operation – June 20, 2013
Basement Temperature Trends - 06/20/13
60
Crawl Space
Near Unit
Dew Point (Near Unit)
40
20
22:
0:
19
20
2:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
21
Dehumidifier Status - 06/20/13
Basement Temperatures and RH -06/20/13
15
Temperature
Run Time (min)
4:
80
100
10
80
60
5
Dehumidifier Set Point: 60% RH
60
0
2022:
0:
2:
4:
6:
8:
10:
Basement Temp (Crawl Space)
Basement Temp (Near Unit)
Basement RH (Near Unit)
12:
14:
16:
18:
20:
19
22:
20
0:
2:
4:
6:
8:
10:
12:
19
20
40
14:
16:
18:
20:
40
22:
0:
22:
21
0:
21
Dehumidifier Status - 06/20/13
Run Time (min)
15
10
5
0
22:
0:
19
20
2:
4:
6:
8:
10:
12:
14:
16:
18:
20:
22:
0:
21
95
20
Relative Humidity (%)
Temperature (F)
80
Figure 79 compares the daily average humidity ratio near the HPWH unit and to the humidity ratio outdoors (using
weather data from the Syracuse Airport). As expected there is a strong correlation between basement humidity and
outdoor humidity due to air leakage. In the winter the humidity in the basement is slightly higher than outdoors due
to moisture generation in the house (HPWH moisture removal in the winter is minimal since the dew point is so
low). In the summer, both HPWH and DH operation tend to remove moisture, which pushes the basement humidity
ratio below ambient.
Figure 79. Comparison of Daily Average Basement and Outdoor Humidity Ratios
Humidity Ratio Comparison
0.015
Basement Humidity (lb/lb)
Winter
0.010
Summer
0.005
0.000
0.000
0.005
0.010
0.015
0.020
Outdoor Humidity (lb/lb)
Although HPWH operation triggers the dehumidifier operation in the short term, on average across a day it reduces
the total load on the dehumidifier by removing some of the moisture from the space. Figure 80 shows the trend of
dehumidifier operation with ambient humidity ratio. The red diamond highlights days when the hot water use was
high and the HPWH operated for more than 600 minutes per day. The green triangles highlight days when the water
use was lower and the HPWH operates less than 300 minutes. The DH operates more when the HPWH is on more
often. The figure also shows that dehumidifier runtime was affected by the outdoor humidity ratio.
96
Figure 80. Effect of Outdoor Humidity Ratio on Dehumidifier Runtime
Outdoor Humidity vs. Dehumidifier Operation
Dehumidifier Runtime (min/day)
1500
1000
HPWH Runtime > 600 min/day
HPWH Runtime < 300 min/day
500
0
0.000
0.005
0.010
0.015
0.020
Outdoor Humidity (lb/lb)
Multi-linear regression used to relate the dehumidifier runtime with HPWH runtime and outdoor humidity ratio.
The relation is given as:
DH
Where DH
=
a0 + a1×WO + a2×HP
- Dehumidifier Runtime (minutes/day)
HP
- HPWH Runtime (minutes/day)
WO
- Outdoor Humidity Ratio (lb/lb)
The detailed statistical results of the multi-linear regression are given in Table 29. The coefficient for HPWH
runtime (a2) has a value -0.215, indicating that each additional hour of HPWH runtime reduces dehumidifier
runtime by 0.215 hours. The t-statistic for this coefficient is greater than 2, indicating it is statistically different than
zero at the 90% confidence level (i.e., significant). This coefficient is used to predict the overall annual impact of
HPWH on dehumidifier runtime and energy use in the table below.
97
Table 29. Multi-linear Regression Results
Parameter
Coefficient
St.Dev.
T-ratio
========================================================
a0
-70.796859
38.188783
-1.8538653
a1
107834.44
3390.6148
31.803802
a2
-0.21532982
0.083295891
-2.5851194
R-Squared
=
0.72706893
Coefficient of Variation, CV =
0.67269096
Mean Bias Error, MBE =
4.5500627e-015
Analysis of Variance
Source
Sum of Squares
D.F.
Mean Square
F-Ratio
===========================================================
Model
1.6060515e+008
3
53535049.
Residual
32952693.
380
86717.612
617.34920
===========================================================
Total
1.9355784e+008
383
Table 30 shows that DH energy is about 864 kWh for the year in this basement. Running the HPWH is estimated to
have reduced DH energy use by 84 kWh over the year, or about 10%. This compares to the $20 savings estimated
by ACEEE (2011) in Section 2.1.7.
Table 30. Monthly Summary of Dehumidifier and HPWH Operation
Apr-13
May-13
Jun-13
Jul-13
Aug-13
Sep-13
Oct-13
Nov-13
Dec-13
Jan-14
Feb-14
Mar-14
Apr-14
May-14
Jun-14
Jul-14
Aug-14
Sep-14
Oct-14
Nov-14
Dec-14
Total
DH
DH
Runtime Energy
(Hrs)
(kWh)
0.0
0.0
0.0
0.0
413.5
124.1
693.3
208.0
625.9
187.8
466.6
140.0
351.5
123.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
232.2
81.3
500.8
150.2
592.3
177.7
739.0
221.7
538.8
161.7
298.0
104.3
0.0
0.0
0.0
0.0
2,783.2
864.1
HPWH
Runtime
(Hrs)
167.3
282.1
228.4
181.8
205.8
176.0
165.2
210.0
250.0
191.5
191.7
245.2
194.0
268.2
182.7
257.3
257.1
147.7
111.3
109.0
146.3
2,507.8
HPWH
Predicted DH
Energy
Reduction
(kWh)
(kWh)
97.0
182.9
139.4
14.7
108.2
11.7
128.9
13.3
101.7
11.4
97.9
12.4
133.5
200.4
120.5
117.5
155.6
125.5
167.0
20.2
112.7
11.8
177.2
16.6
155.4
16.6
277.2
9.5
294.5
8.4
290.0
418.3
1,596.1
83.7
Note: The highlighted months are months before the dehumidifier status sensor was installed
98
5.6.3 Impact of Recirculation Pump on System Performance
On April 10, 2013 the distribution system at Site 1 was modified to include a Bell & Gossett ecorcirc pump (the
same time the HPWH was installed). This recirculation pump includes a built-in time clock to initiate operation as
well as a built-in temperature sensor to stop operation once return water at the pump has warmed to the desired set
point (the setting was 85°F). The pump was installed near the tank and pulls water from supply lines at two remote
locations: kitchen sink and half bath sink. Half inch PEX was used for the recirculation line (see Appendix A).
The pump was scheduled operate at times when hot water use in this home was considered probable (i.e., 5:30 to
7:30 AM, 10:30 to 11:30 AM, and 3:30 to 5:30 PM each day, EST). All total, the recirculation pump is allowed to
operate for up to 5 hours each day. However, the actual runtime was typically in the range of 2–6 minutes/day
because of the pump’s controller sensed that the return water was already at the set point of 85F.
Figure 81 is a shade plot that qualitatively shows when the pump operated. Each day is shown as a vertical stripe on
the plot with shades of gray. Subsequent days are shown along the x-axis. Periods when the pump operates are
shown with darker shades of gray. The shade plot shows that the pump always operates at the beginning of each of
the three periods. Operation later in the period is intermittent.
Figure 81. Shade Plot Showing Runtime of Recirculation Pump
Recirc Pump
24
22
20
18
Hour of Day
16
14
12
10
8
6
4
2
0
April
May
2013
Day (MAX/MIN =
2.33/ 0.00 )
99
The evaluation from the Building America report (Henderson and Wade 2014) showed that the useful portion of the
delivered hot water increased from 82% to 91% after the pump was installed. Assuming an average water use of 48
gallons per day this equates to reduction in water use of 4 gallons per day. The energy required by the HPWH to
heat this saved water is about 0.4 kWh/day (assuming a conversion efficiency of 1.5 for the HPWH), and the
standby losses for the installed HPWH with the pump operating was about 2 kWh/day. In contrast the same HPWH
in the lab required 0.6 kWh/day to maintain the tank temperature with no water use. So recirculation pump
operation reduces hot water use but increases standby losses. The net impact of these two factors is summarized
below.
Savings from reduced hot water use (4 gal/day less)
0.4 kWh/day
Additional Standby losses (2 kWh/day minus 0.6 kWh/day)
1.4 kWh/day
Net savings
-1.0 kWh/day
Operating of the recirculation pump did NOT result in net energy savings. Energy use increased by about one kWh
per day.
Before the HPWH was installed the indirect tank on the oil-fired boiler met the water heating load. Standby losses
for the indirect tank resulted in an average boiler runtime of 9.4 minutes for each day with no water use. The fuel oil
boiler nozzle and pressure resulted in a 0.89 gallon per hour rating. Assuming 0.89 gal/h x 140 MBtu/gal x 9.4/60 *
0.85 the thermal losses are 16.5 Btu/h on average.
5.7
Field Test Site 2 – Condensing Storage Tank
At Site 2 a new AO Smith Vertex (GDHE-50) condensing storage tank water heater was installed near the location
of the original water heater on May 21, 2013. At the same time a Taco SmartPlus® recirculation pump was installed
on the supply line with the sensor downstream of the pump. A ½-in PEX return line was added from the supply line
under the kitchen sink back to the cold water inlet to the water heater. The pump, combined with the return line,
allowed the first 13 ft of supply piping towards the kitchen to stay at temperature in order to reduce hot water waste.
The pump was set in the “Smart/Learn Mode,” where it ran based on the pattern established over the last 7 days of
operation. In this Smart Mode, the pump runs in the Pulse Mode (running for 2.5 minutes out of every 10 minutes)
for a 2-hour window centered around each draw recorded from one week (168 hours) ago.
Figure 82 is a shade plot showing when the recirculation pump operated. Each day is shown as a vertical stripe on
the plot. 15-minute intervals with no pump operation are shown as light gray. Periods with 2.5 minutes of operation
are shown as medium gray. Darker periods have more runtime in the 15-minute interval. Since the hot water usage
100
pattern in the house was irregular, the pump controls did not settle into a consistent weekly pattern. The most
consistent pattern may have been for few weeks in November.
Figure 82. Shade Plot Showing When Recirculation Pump Operated
Recirculation Pump Runtime
24
22
20
18
Hour of Day
16
14
12
10
8
6
4
2
0
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Nov
Dec
2013
Day (MAX/MIN =
5.08/ 0.00 )
Figure 83. Plots of Daily Runtime for Recirculation Pump
Runtime (min/day)
400
200
0
Jun
Jul
Aug
Sep
Oct
2013
2014
101
Figure 83 shows how the daily total runtime varied over the same period. The measured runtime for the pump was
about 200 minutes/day on average (about 14% of the time). The “Smart/Learn Mode” reduced the runtime by only a
modest amount compared to the “Pulse Mode” (which would have run the pump 25% of the time 24 hours/day). In
contrast, the timer-controlled pump at Site 1 runs the pump only about 3 minutes/day in the 5-hour window when
operation is enabled.
Recirculation pump operation was disabled on December 2, 2013. Figure 84 compares the gas use verses hot water
use for similar periods with and without the pump operating. The days shown in the plot all had incoming cold
water temperatures between 51°F and 56°F. Operation of the recirculation pump (black data) showed increased
thermal losses (gas use) compared to periods without the pumping operating (red data). The data in Table 31 show
the y-offsets for the best fit lines to these two data sets. The operation of the recirculation pump increased losses by
5 MBtu/day, from 27 MBtu/day to 32 MBtu/day. In both cases standby losses were much higher than the 7
MBtu/day that had been measured in the laboratory.
Figure 84. Impact of Recirculation Pump Operation on Water Heater Gas Use (when cold water is
between 51 and 56°F)
DHW2 - 05/21/13 to 05/25/14
200
Gas Use (MBtu/day)
150
100
Recirculation Pump Off
50
Laboratory Losses = 7 MBtu/day
0
0
50
100
150
Water Use (gal/day)
102
200
Table 31. Impact of Recirculation Pump Operation on Standby Losses
Standby Losses
(gas use no water use)
Laboratory Measurements
7 MBtu/day
Recirculation Pump Operating
May 21 to December 2, 2103
32 MBtu/day
No Recirculation Pump
December 3, 2013 to May 25, 2014
27 Mbtu/day
Figure 85 shows the measured conversion efficiency verses hot water use. The efficiency was 0.4 to 0.6 – which is
much lower than the 0.86 that was measured in the laboratory for the standard water use profile. At least some of
the scatter in the data is due to the changes in water temperatures across the year. Although other unknown thermal
loss mechanisms may have also played a role given the much the much lower than expected conversion efficiency.
Figure 85. Variation of Efficiency with Hot Water Use for Period With Recirculation Pump Off
DHW2 - 05/21/13 to 05/25/14
1.0
Daily Conversion Efficiency (-)
Laboratory EF is 0.86
@ 64 gal/day
0.8
0.6
Recirculation Pump Off
0.4
0.2
0
50
100
150
200
Water Use (gal/day)
103
250
300
5.8
Field Test Site 3 – Gas-Fired Condensing Tankless
At Site 3 a tankless water heater (Rinnai RL75i) was installed to replace the conventional gas-fired tank on March
22, 2013. Details about the water heater and installation are given in Appendix A. The impact of this change on this
household with two people is shown in the plots below. Figure 86 shows data for the pre-retrofit period with the
conventional gas-fired tank as black, while data with the tankless unit (after March 22) are shown as red. Total water
use changed only slightly; however, the number of hot water draws was reduced by more than 50% with the tankless
unit. The minimum activation flow threshold of 0.4 gpm for the tankless unit (as well as the startup delay) caused
the occupants to change their hot water use.
The evaluation from the Building America Report (Henderson and Wade 2014) showed that the useful portion of the
delivered hot water decreased from 90%–91% with the conventional tank to 84%–86% with the tankless unit. This
4-6% reduction in useful delivered hot water (about 1.7 gallons per day) was mostly due to the startup delay of the
tankless unit (i.e., the time delay for burner activation).
Figure 86. Comparing HW Use and Events for Conventional Tankless Systems
Site 3 (excluding vacations)
AVG = 34.5
AVG = 32.0
Water Use (gal/day)
120
100
80
60
40
20
0
31 7 14 21 28 4 11 18 25 4 11 18 25 1 8 15 22 29 6 13 20 27 3 10 17 24 1 8 15 22
January
February
March
April
May
June
Draw Events (no/day)
AVG = 52.3
July
AVG = 20.1
150
100
50
0
31 7 14 21 28 4 11 18 25 4 11 18 25 1 8 15 22 29 6 13 20 27 3 10 17 24 1 8 15 22
January
February
March
April
May
104
June
July
The conversion efficiency of the tankless was typically 0.75, just below the rated EF of 0.82. As expected the
tankless unit has little or no standby losses, so the conversion efficiency is not strongly dependent on daily water
use.
Figure 87. Conversion Efficiency Versus Hot Water Use
03/22/13 to 04/02/15
1.0
Rated EF is 0.82 @ 64
gal/day
Conversion Efficiency (-)
0.8
0.6
0.4
0.2
0.0
0
20
40
60
Hot Water Use (gal/day)
105
80
100
6
Conclusions and Lessons
6.1
Summary of Findings and Conclusions
This study evaluated a wide array of issues related to the performance of water heaters in New York homes. The
study looked at the performance of water heater products ranging from standard gas water heater tanks through high
efficiency gas-fired options. It also looked at solar and heat pump options as an alternative to standard electric water
heaters. We used laboratory testing to make comparisons between different systems and to discern the impact of the
new water use profiles in the forthcoming DOE Test Procedure to determine the new Uniform Energy Factor (UEF).
Because the water use pattern can have a significant impact on performance, we also measured hot water flow in
several homes to understand the magnitude and timing of hot water use. Data were also collected to quantify hot
water waste due to the distribution system in several homes. Field testing also provided the means to measure the
performance of water heating systems under actual, in-situ conditions.
6.1.1 Impact of Draw Profile on Efficiency
Residential water heaters are currently rated by the Energy Factor (EF) which is determined by a simulated use test
(SUT) procedure that imposes a hot water draw pattern on the appliance and then measures the thermal energy
output and energy input over a 24 hour cycle. One issue with the current procedure is the unrealistic water draw
profile that included six large draws taken one hour apart for a total 64.2 gallons per day. After several years of
rulemaking with the industry, DOE recently modified this procedure (July 11, 2014) to include more realistic water
use patterns. The low, medium or high use patterns are used in the test procedure depending on the first hour rating
(FHR) of unit. The updated test rating procedure determines a new descriptor known as the Uniform Energy Factor
(UEF).
We imposed the original (or current) six-draw use profile on the water heaters in the laboratory and then also
imposed the proposed multi-event patterns from ASHRAE 118.2, which included low use (40 gallons per day),
normal use (64.2 gallons per day) and high use (82.8 gallons per day).
Generally we found that the standard gas-fired water heater (Gas STD) and the high efficiency power vented water
heater (HE-PVNT) have conversion efficiencies that were very close to the rated EF. The conversion efficiency was
only slightly affected by the more realistic normal draw profile.
For the tankless condensing water heater (TANKLESS) we found a conversion efficiency slightly lower than its EF,
or 0.90 compared to 0.94 EF. Going from six large water draws to the 12 smaller draws at the same daily usage
level slightly lowered the conversion efficiency from 0.9 to 0.86. The lower use profile resulted in an even lower
conversion efficiency (dropping from 0.86 to 0.81) and the high use profile resulted in a higher efficiency
106
(increasing fron 0.86 to 0.88). Therefore, the new UEF is likely to be more representative of actual performance for
tankless water heaters. The large draws of the current EF procedure overstated actual efficiency.
The high-efficiency, condensing gas-fired tank unit (HE-Cond) and the Hybrid unit (which is a tankless unit with a
small internal storage tank) are both currently rated by thermal efficiency (TE) under a test intended for commercial
water heaters. For both these units, the measured conversion efficiency was lower than the TE, as would be
expected. Going forward these units will be rated under the new UEF descriptor. For HE-Cond unit the conversion
efficiency was 0.86 compared to 0.96 TE. Field testing by the manufacturer had shown efficiencies of 0.81 for this
unit (Adams 2008). For the HYBRID unit the conversion efficiency was 0.73 compared to 0.90 TE. The efficiency
did vary considerably for the HYBRID unit depending on the water use profile. The conversion efficiency varied
from 0.65 at low usage to 0.78 for high usage. The strong dependence on usage is even greater than for the tankless
unit.
The measured conversion efficiency of 2.53 for the HPWH in the laboratory was slightly higher than the 2.4 EF with
conventional six draw profile. The warmer conditions in the lab (75°F) compared to the test procedure (68°F)
contributed to the better-than-expected performance. The low use profile did result in a slightly lower conversion
efficiency as would be expected. The high use profile caused the resistance elements to operate to recover from the
large draws in the pattern. This lowered the overall conversion efficiency to 1.77, or by 25% compared to the
normal use efficiency.
6.1.2 Side-by-Side Solar Testing
Two solar systems were tested side-by-side in the laboratory. The first system was a conventional 120 gallon single
tank unit (SOLAR) with an internal heat exchanger coil connected to pressurized glycol loop that ran to the solar
collector on the roof. This system is one of the most common solar products sold in New York State. The second
system was a new glycol drain-back system (SOLAR-Drain) designed to provide the widely-recognized
performance benefits of drain-back systems to a cold climate application. This two tank system included a 92 gallon
solar pre-heat tank with internal heat exchanger that could hold the entire charge of glycol when the collector pump
was off. The single tank conventional system had an OG-300 Solar Energy Factor (SEF) of 3.2. The expected
annual Solar Fraction (SF) from SRCC for Syracuse was expected to be 0.56. We measured a solar fraction of 0.47
for the year, which generally in-line with expectations.
The drain-back system initially had performance problems where it would vapor-lock and stop operating on sunny
days. Once this problem was corrected the system behaved as expected, and we saw slightly better performance for
the drain-back system on partially sunny days. However, the improved performance at part load conditions did not
translate into an overall improvement compared to the baseline solar system. The annual solar fraction for the drainback system was just slightly less than the baseline one-tank system at 0.46, in spite of the expected performance
benefit of having two tanks.
107
6.1.3 Hot Water Use Patterns in Test Homes
We measured hot water use in 18 single family homes around Syracuse. The median water use for all the homes
was 45 gal/day with an average occupancy of 2.7 people per household. The median for the sample of 2-person
households was 39 gal/day, which is in-line with larger national studies such as Parker (2013) which estimates 38
gal/day and a study in the Pacific Northwest (Larson 2015) that found 34 gal/day. The median for the 4-person
households was 66 gal/day compared to 63 gal/day nationally (Parker 2013) and 56 gal/day for the Northwest
(Larson 2015).
Hot water use in a given household varies significantly from day to day. For instance, for the 2-person households
the standard deviation of water use within each home ranged from 8 to 23 gal/day, compared to a median of 39
gal/day. So peak water use is often 2 to 3 times median water use.
We also noted seasonal variations in water use, presumably as the inlet water temperature changed. The houses on
public water supply systems – where water pipes are often close to ground level – showed much more variation for
the inlet water temperature across the year. Inlet water temperatures typically ranged from 38°F to 70°F across the
year. For rural houses on wells the inlet temperature varied much less (48°F to 60°F).
The degree of seasonal water use variation for the 2-person households was much smaller than for the 4-person
households. This is mostly likely due to the greater importance of showers in mix of water use activities for
households with more people. During a shower, occupants clearly mix in more hot water in the winter months to
achieve the desired outlet temperature at the showerhead. For other water uses, the volume of hot water may be less
sensitive to cold water temperatures.
We developed average profiles for water use based on the 15-minute data. The average profiles showed a use
pattern that is biased towards morning use for the 2-person household. For the 4-person households the profile was
more evenly divided between morning and evening use – perhaps because the occupants modify their behavior and
schedule to accommodate the needs of the entire household.
At 13 of the test homes we also collected detailed data about each water use event. Water use patterns are more
realistically represented as a series of distinct draw events across a day instead of the continuous daily profiles
determined from the 15-minute data. The size and timing of draw events are what affect system performance (i.e.
large draws can cause resistance element operation for the HPWH). While there is significant variation between
sites we found that the average draw event was about one gallon – so a house that uses 50 gallons per day will
typically have 50 draw events across the day.
108
6.1.4 Disaggregating Hot Water Use and Estimating Hot Water Waste
At five of the field test sites we installed temperature sensors on the distribution piping near each fixture and on
some trunk lines to allocate water draws to each end use and to estimate the amount of hot water waste. This
analysis technique was only able to classify about half of the draw events across the five sites. However, it was able
to account for 95% of hot water use at the sites. The large number of unclassified events was typically very small
draws.
Using the temperature sensors on the pipe surface near each fixture, we were able to estimate the amount of
unusable previously-heated water that was run down the drain before useful hot water reached each fixture. We
selected a threshold temperature of 90°F at the pipe surface to correspond to an acceptable hot water temperature at
the fixture. Using these criteria we found that from 9% to 25% of the hot use at each site was associated with hot
water waste – i.e., previously-heated water that is run down the drain waiting for an acceptable temperature at the
fixture. This matches well to the estimated waste of 2 gallons per day per occupant from Parker (2013), which
equates to about 11% hot water waste. Hot water waste was found to be highest at kitchen and bathroom sinks and
lowest for showers and washing machine draws.
6.1.5 HPWH Field Performance
A HPWH was installed at Site 1 in a wet, unconditioned basement near a hot water boiler. The HPWH, which was
the same as the laboratory test unit, was found to have an overall annual conversion efficiency of 1.4 while operating
in the hybrid mode. Resistance element operation on days with large water use events (e.g., draws over 20 gallons)
resulted in the lower efficiency. The controls on the HPWH unit required full recovery of tank temperature with the
resistance element. Cold air temperatures in the basement (50-65°F) also reduced efficiency.
The HPWH unit operated in the resistance-only mode for significant portion of the test period (due to a problem
with compressor). This provided the opportunity to directly compare hybrid heat pump operation and resistanceonly operation for the same tank in the same distribution system. The efficiencies in these two modes were different
by a factor of two, mainly because the configuration in this site resulted in a conversion of efficiency of the unit in
the resistance mode of 0.67. While thermal standby losses from the HPWH were probably higher than for a
conventional electric tank, this comparison still implies that the HPWH reduces electric use compared to a standard
electric by nearly half.
The HPWH operation had only a moderate impact on space conditions, lowering basement air temperatures by 24°F on average. Therefore, there was no discernible impact of HPWH operation on the heating load in the
conditioned space above. The basement also had a dehumidifier that kept the basement below 60% RH in the
summer. From the measured data we determined that each 5 hours of HPWH operation reduced dehumidifier
109
runtime by one hour, reducing dehumidifier energy use by about 84 kWh annually. The savings of $10 per year are
slightly less than the $20 dehumidification benefit estimated by ACEEE (2011).
6.1.6 Impact of Switching to Tankless Unit
At Site 3 a standard gas water heater was monitored for a few months and then replaced with a non-condensing
tankless unit with an EF of 0.82. Installing the tankless unit did not significantly change the daily average water use
of the household, which was about 34 gallons per day. However, it did change the number of hot water draw
events. With the tankless unit the number of draws was reduced by more than half, from 54 to 20 draw events per
day. The occupants adjusted their behavior presumably because small draws no longer provided hot water at the
desired temperature because of the time delays associated with burner startup in the tankless unit. On average the
conversion efficiency was about 0.75 at 64 gallons per day, slightly below the rated EF.
6.1.7 Field Performance of High Efficiency Condensing Tank
At Site 2 we installed the same condensing storage tank that was installed in the laboratory. The condensing storage
tank had much lower efficiency then was measured in the laboratory. The measured conversion efficiency of 0.5 at
64 gallons per day compared to 0.86 in the laboratory. Part of the poor performance was due the use of a
recirculation pump at this site. The recirculation pump reduced hot water waste but greatly increased thermal
standby losses from this system.
6.1.8 Recirculation Pumps to Reduce Hot Water Waste
At two field test sites, we installed different types of recirculation pumps along with a return line back to the tank
inlet. The pumps used slightly different control methods to keep the main supply trunk primed with hot water to
reduce wait times and hot water waste. At both sites the pumps reduced hot water waste but had an even greater
impact on standby losses. Hot water waste was reduced by about 4 gallons per day at Site 1 due to pump operation,
but standby losses were increased by more than the savings from eliminating this waste. At Site 1 we calculated a
net increase in energy use of 1 kWh per day. Recirculating water clearly increases thermal losses from the supply
and return piping near the tank; so while water use is lower, energy use increases. Therefore in regions like the
Northeast where energy savings are of more concern with than water conservation, recirculation pumps do not seem
to be justified.
6.2
Lessons Learned
6.2.1 One Water Use Profile May Not Fully Characterize Performance
The new simulated use test procedure to determine the UEF rightly selects an appropriate usage profile (low,
medium or high) based on the first hour rating (FHR) or “size” of the water heater. However, measured field data
110
show that water use in a given house can vary significantly from day to day as the occupants engage in different
activities. Therefore, appliances such as the HPWH will have efficiencies approaching the UEF on days when usage
in near normal, but much lower efficiency on high use days when the resistance element is engaged. In this respect,
the new test procedure still may fall short of representing actual efficiency.
6.2.2 Controls and Tank Size Matter for HPWH
The performance of the HPWH we tested in laboratory and field demonstrated the importance of control in
providing efficient operation. In the hybrid mode, the unit controls put priority on using the resistance elements to
provide a rapid recovery after a large draw. Obviously, this control strategy is important in achieving the first hour
rating (FHR) for the unit. Control approaches that consider time of day and others factors when deciding when and
how much to use the resistance elements versus the heat pump would greatly impact actual efficiency. Evaluations
at NREL (Sparn et al. 2014) have highlighted the benefit of improved control strategies on HPWH efficiency.
Many manufacturers are now on their second and third generation of HPWH products. A NYSERDA-sponsored
field test of different first and second generation HPWH units in Delaware County is expected to show the
improvements in performance for different generations of these products (Ealey 2015). In addition to controls,
many manufacturers now have units with larger storage tanks to minimize the chance of element operation during a
large draw. For instance General Electric now offers both a 50 gallon and an 80 gallon version of their HPWH.
6.2.3 HPWH is an Energy Efficient Option in Rural Homes
There are 1.2 million New York homes that use electric resistance water heaters. Another 1.3 million homes use oil
for water heating. Several hundred thousand of these households are rural and suburban single family homes with
basements which are good candidates for a HPWH installation. The HPWH offers significant savings compared to
electric and even offer savings compared to oil-fired water heating options. These units are now competitively
priced at big box stores, implying a strong do-it-yourself market is developing for these relatively easy-to-install
units.
111
7
References
ACEEE. 2011. Emerging Hot Water Technologies and Practices for Energy Efficiency as of 2011. Report A112.
American Council for an Energy Efficient Economy. Washington, DC. Prepared October 2011. Revised
February 2012.
Adams, C. 2008. “A. O. Smith Vertex Condensing Tank-type Water Heater.” ACEEE Hot Water Forum
Sacramento. June 3.
AHRI. 2011. Directory of Certified Product Performance. Retrieved August 15, 2011, from AHRI Directory for
Water Heaters: http://www.ahridirectory.org/ahridirectory/pages/rwh/defaultSearch.aspx
ARHI. 2013. MODIFIED SIMULATED USE TEST FOR DETERMINING ENERGY DESCRIPTOR. March.
Draft from an Standard ASHRAE 118.2 meeting.
C. Dennis Barley; R. Hendron; Lee Magnusson. 2010. “Field test of a DHW distribution system: temperature and
flow analyses” Presented at ACEEE May 2010 Hot Water Forum. NREL/PR, 550-48385.
DOE. 2011. Energy Savers. from Energy Efficiency & Renewable Energy:
http://www.energysavers.gov/your_home/water_heating/index.cfm/mytopic=12840 February 9. Retrieved
August 17, 2011,
DOE. 2010. Energy Star Water Heater Market Profile 2010. U.S. Deparment of Energy, Energy Star. Prepared by
D&R International, Ltd. For Oak Ridge National Lab.
http://www.energystar.gov/ia/partners/prod_development/new_specs/downloads/water_heaters/Water_Heater_
Market_Profile_2010.pdf
Ealey, B. 2015. “Heat Pump Water Heater Demonstration at Delaware County Electric Cooperative Delhi, NY.
ACEEE Hot Water Forum, Session 1C. Nashville. February.
EIA. 2013. Residential Energy Consumption Survey (RECS) for 2009 from U.S. Energy Information
Administration. Household Site End-Use Consumption in the Northeast Region, Totals and Averages, 2009.
Table CE3.2. http://www.eia.gov/consumption/residential/data/2009/c&e/end-use/xls/CE3.2 Site End-Use
Consumption in Northeast.xlsx
EIA. 2013b. Residential Energy Consumption Survey (RECS) for 2009 from U.S. Energy Information
Administration. Fuels Used and End Uses in Homes in Northeast Region, Divisions, and States, 2009. Table
HC1.8. http://www.eia.gov/consumption/residential/data/2009/xls/HC1.8 Fuels Used and End Uses in
Northeast Region.xls
112
EIA. 2013c. Residential Energy Consumption Survey (RECS) for 2009 from U.S. Energy Information
Administration. Water Heating in U.S. Homes in Northeast Region, Divisions, and States, 2009. Table HC8.8.
http://www.eia.gov/consumption/residential/data/2009/xls/HC8.8 Water Heating in Northeast Region.xls
EIA. 2013d. Residential Energy Consumption Survey (RECS) for 2009 from U.S. Energy Information
Administration. Space Heating in U.S. Homes in Northeast Region, Divisions, and States, 2009. Table HC6.8.
http://www.eia.gov/consumption/residential/data/2009/xls/HC6.8 Space Heating in Northeast Region.xls
Energy Star. 2011. Energy Star Criteria Water heater, Heat Pump
http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductGroup&pgw_code=WHH
Hoeschele, M. and L. Brand. 2013. “Hot Water in California: Single Family Perspective.” CEC Water Heating
Workshop, July 16.
Klein, G. 2011. “Plenary Session - Where is Hot Water in High Performance Buildings” Presented at ACEEE May
2010 Hot Water Forum, Berkeley, CA.
Klein, G. 2006? “Saving Water and Energy in Residential Hot Water Distribution Systems” Slides from California
Energy Commission at DOE website, (2005)
Larson, B. 2015. “Highlights from a Field Test Study of HPWHs at 100+ Houses Across the Pacific Northwest.”
ACEEE Hot Water Forum, Session 1C. Nashville. February.
Lutz, J. 2009. “Estimating Energy and Water Losses in Residential Hot Water Distribution Systems”. Lawrence
Berkeley National Laboratory. Paper LBNL-57199
Magnusson, L. 2009. “Methods and results for measuring hot water use at the Fixture National Renewable Energy
Lab. Presented at ACEEE June 2009 Hot Water Forum
Neale, A. 2013. “Results from a Canadian Hot Water Pilot Project.” ACEEE Hot Water Forum, Session 7D.
Atlanta, GA. November.
NEEA. 2011. A Specification for Residential Heat Pump Water Heaters Installed in Northern Climates Version
4.0. Northwest Energy Efficiency Alliance. http://neea.org/docs/northern-climate-heat-pump-water-heaterspecification/northern-climate-specification.pdf?sfvrsn=4
NYSERDA. 2014. Patterns and Trends New York State Energy Profiles: 1998-2012. New York State Energy
Research and Development Authority. Albany, NY. November.
Parker, D. 2013. “Using Occupancy to Develop an Improved Predictive Model of Residential Hot Water
Consumption.” ACEEE Hot Water Forum, Session 7D. Atlanta, GA. November.
Sparn, B. K. Hudon and D. Christensen. 2014. “Laboratory Performance Evaluation of Residential Integrated Heat
Pump Water Heaters.” Technical Report NREL/TP-5500-52635. Golden. Revised June.
113