Information about http://enduse.lbl.gov/Info/LBNL-35475.pdf

…

Tags: berkeley ca, contract number, d lutz, energy analysis program, energy consumption, energy use, environment division, equipment efficiency standards, existing equipment, heating energy, heating water, koomey, lawrence berkeley laboratory, office of energy efficiency, primary energy, residential sector, residential water, resource planners, showerheads, water heating,
Pages: 21
Language: english
Created: Fri Sep 20 10:21:27 1996

Display cached document

Page 1

Page 2

Page 3

Page 4

Page 5

Page 6

Page 7

Page 8

Page 9

Page 10

Page 11

Page 12

Page 13

Page 14

Page 15

Page 16

Page 17

Page 18

Page 19

Page 20

Page 21

LBL-35475
UC-000

The Effect of Efficiency Standards on Water Use and Water Heating
Energy Use in the U.S.: A Detailed End-use Treatment
Jonathan G. Koomey, Camilla Dunham, and James D. Lutz

Energy Analysis Program
Energy and Environment Division
Lawrence Berkeley Laboratory
Berkeley, CA 94720

May 1994

The work described in this paper was supported by the Office of Energy Efficiency and Alternative
Fuels Policy, Office of Policy of the U.S. Department of Energy under Contract Number DE-AC03-
76SF00098.
The Effect of Efficiency Standards on Water Use and Water Heating Energy Use in
the U.S.: A Detailed End-use Treatment1

Jonathan G. Koomey, Camilla Dunham, and James D. Lutz
Lawrence Berkeley Laboratory

SYNOPSIS

This paper analyzes US residential water use and water heater energy use. It also
investigates water and energy savings associated with existing equipment efficiency
standards.

Keywords: Residential codes and standards, Water heating, Water conservation,
Forecasting, Program design

ABSTRACT

Water heating is an important end-use, accounting for roughly 16% of total primary energy
consumption in the US residential sector. Recently enacted efficiency standards on water
heaters and hot water-using equipment (e.g., dishwashers, clothes washers, showerheads,
and faucets) will substantially affect the energy use of water heaters in the future.
Assessment of current and future utility programs and government policies requires that
regulators, resource planners, and forecasters understand the effects of these regulations.

In order to quantify these impacts, this paper presents a detailed end-use breakdown of
household hot and cold water use developed for the US Department of Energy. This
breakdown is based on both previous studies and new data and analysis. It is implemented
in a spreadsheet forecasting framework, which allows significant flexibility in specifying
end-use demands and linkages between water heaters and hot water-using appliances.

We disaggregate total hot and cold water use (gallons per day) into their component parts:
showers, baths, faucets (flow dominated and volume dominated), toilets,
landscaping/other, dishwashers, and clotheswashers. We then use the end-use breakdown
and data on equipment characteristics to assess the impacts of current efficiency standards
on hot water use and water heater energy consumption.

1 A shortened version of this report will be published in the proceedings to the 1994 ACEEE Summer
Study Conference on Energy Use in Buildings.

i
ii
TABLE OF CONTENTS

SYNOPSIS
ABSTRACT
1. INTRODUCTION ....................................................................................................1
2. REVIEW OF CORE DATA ON WATER USE AND WATER
HEATING ENERGY USE ..........................................................................................1
2.1 Characteristics of water heaters...............................................................1
2.2 Water temperature assumptions...............................................................3
2.3 Showers, sinks, and faucets...................................................................... 3
2.4 Toilets....................................................................................................... 6
2.5 Dishwashers and clotheswashers ............................................................6
2.6 Other uses ................................................................................................6
3. FORECASTING METHODOLOGY AND RESULTS......................................6
3.1 Forecasted number of households ...........................................................6
3.2 Baseline and standards case forecasts: water use.................................. 9
3.3 Baseline and standards case forecasts: energy use ................................9
3.4 Energy savings.........................................................................................9
4. FUTURE WORK......................................................................................................9
4.1 Sensitivity analysis/further disaggregation............................................... 9
4.2 Energy losses in delivering water to the tap............................................. 13
4.3 Measurement of water use........................................................................13
4.4 Comparison to other forecasts.................................................................13
4.5 "Upstream" and "Downstream" issues....................................................13
4.6 Enforcement of standards .......................................................................13
5. CONCLUSIONS.......................................................................................................13
ACKNOWLEDGEMENTS.........................................................................................14
REFERENCES..............................................................................................................14

iii
iv
1. INTRODUCTION

The water heating end-use is perhaps the most complicated of all residential end-uses.
Usage behavior varies greatly, and depends on ownership of other appliances (e.g., clothes
washers and dishwashers), on the characteristics of those appliances and other water-using
equipment, on water inlet temperatures, on hot water temperature set points, and on water
use temperatures.

Water heating has generally been analyzed in a haphazard manner in the past, in large part
because of its complexity. Measurement studies (with a few notable exceptions) have
failed to capture the important data necessary for truly understanding how hot and cold
water are used in US residences. For example, studies often report the number of gallons
per day of hot water use per household, averaged over all households with different
appliances. This number is only useful for determining total water heating energy use�it
provides no guidance as to what impact different conservation measures might have on that
usage. It is also of limited usefulness in assessing the energy use for water heating in a
particular household. The only way to assess such issues is to disaggregate total hot water
use to reflect the various components of hot water loads.

The guiding philosophy behind this paper is that explicit and well-chosen estimates about
usage behavior for particular devices (e.g. the number of minutes each person showers per
day) can be used to create a "bottom-up" estimate of total usage that is a reasonable
portrayal of average water usage. Such detailed bottom-up estimates can be used to assess
the effects of policies. In the case of water heating, creating such a disaggregated forecast
is the only way to disentangle the effects of the various efficiency standards now affecting
this end use. The "order of magnitude" character of the numbers is acceptable for most
applications as long as they are derived in an internally consistent and complete manner,
and as long as available data, basic physical principles, and common sense are used to
guide the choice of parameters.

Characteristics of particular houses may be quite different from the usage assumptions
shown here. Nevertheless, we believe that our first-order estimates crudely represent
average water usage behavior in US residences. We eagerly await more detailed data on
water usage from the many studies now being conducted by US utilities and others.

2. REVIEW OF CORE DATA ON WATER USE AND WATER HEATING
ENERGY USE

2.1 Characteristics of water heaters

Table 1 shows the characteristics of residential water heaters of various vintages, derived
from US DOE (1993). The relevant characteristics are energy factor (defined as energy
delivered in the hot water at the tap, divided by total energy input), recovery efficiency
(which is the efficiency of delivering heat to the water in the tank), and the standby losses
(which are the constant losses associated with keeping the water in the tank hot and ready
for use).

The energy factor, recovery efficiency, and standby losses are related using Equation (1):

Energy content of hot water delivered
EF = (1)
Total energy used to heat the water

1
Table 1: Characteristics of water heaters

Stock Stock
Units 1990 1993 Sales Sales Proposed Stds
w/1990 stds late 1980s 1990 1-Jan-98
Energy Factors
Electricity % 82% 83% 84% 88% 189%
Gas % 49% 50% 49% 54% 57%
Oil % 46% 46% 47% 51% 67%

Standby losses
Electricity Wh.th/hr 97 89 86 57 10
Electricity Btus.th/hr (site) 332 303 294 194 34
Gas Btus.th/hr 1576 1453 1510 1125 948
Oil Btus.th/hr 1576 1453 1510 1125 948

Recovery efficiency
Electricity % 98% 98% 98% 98% 193%
Gas % 76% 76% 76% 76% 76%
Oil % 76% 76% 76% 76% 76%

(1) Standby losses and recovery efficiency for gas and electric water heaters (WHs) derived
from US DOE (1993). Oil water heaters assumed to be the same as gas.
(2) Energy factors (EFs) for stock (1990 and 1993) and late 1980s sales derived from
historical shipments taken from Hanford et al. 1994. EFs for 1990 sales equal the 1990 standards.
Proposed 1998 standard on electric WHs requires manufacturers to produce only heat pump water heaters.
Recovery efficiency for heat pump water heaters varies depending on water temperatures.
Recovery efficiency shown here is an overall average.
(3) (Standby losses + the heat content of water used)/recovery efficiency = total WH energy use.
Energy factor = the heat content of water used/total WH energy use.
(4) Calculation of Stock and late 1980s sales standby losses assumes that all measures used by
manufacturers to improve the WH EF affected only standby losses.
(5) Wh.th/hr = watt-hour (thermal) per hour; Btus.th/hr = Btus (thermal) per hour.
Where total energy used to heat the water (Etotal) is defined as in Equation (2):

(Energy content of hot water delivered + Standby losses)
Etotal = (2)
Recovery efficiency

We use these equations to create the characteristics of electric and natural gas water heaters
in Table 1, using data from US DOE (1993). Energy factors (EFs) are given either from
historic data (Hanford et al. 1994) or for the 1990 standards. These energy factors are used
with the above equations to calculate standby losses and recovery efficiencies for cases
where those parameters are not immediately available. The calculation of "Stock" and
"Late 1980s sales" standby losses assumes that all measures used by manufacturers to
improve the water heater EFs affected only standby losses.

2.2 Water temperature assumptions

Table 2 describes the end-use temperatures assumed in our calculations. The hot water
temperature is the weighted average from a survey of water heater temperatures cited in
Bancroft et al. (1991). The cold water temperature represents a rough average across the
US and over the year, adopted from data in Labs (1979). These two temperatures and the
assumed end-use temperatures imply a certain percentage breakdown of hot and cold water
use. These percentage breakdowns are used to calculate hot and cold water use from total
water use below.

2.3 Showers, sinks, and faucets

Table 3 summarizes usage times and flow rates for showers, baths, and faucets. We
disaggregate these uses into flow and filling uses because the standards have different
effects on them. Flow rate standards affect only the flow portion of the water use. Filling
water uses are not affected by these flow rate standards. Flow assumptions for these end-
uses were created after reviewing the available literature (Bancroft et al. 1991, Kempton
1984, Perlman 1987, Rohles Jr. and Konz 1981, Taylor et al. 1991, Thrasher et al. 1990).
We rely most heavily on Brown and Caldwell (1984), which has the most careful and
comprehensive treatment of total water use data (though its treatment of hot and cold water
disaggregation is cursory at best).

Flow rates for existing showerheads (and faucets) are 3.4 gallons per minute, which is
approximately the flow rate found by Brown and Caldwell (1984) and Warwick (1993).
The EPACT standards reduce flow rates to 2.5 gallons per minute. Shower usage is
assumed to be 5 minutes per person per day (2.67 persons per household), which is again
based on Brown and Caldwell (1984), and is on the high end of estimates from Rohles et
al. (1981) Faucet filling is always assumed to be equal to base case flow for bathrooms,
kitchens, and laundry faucets, respectively (filling is not affected by standards).

For homes without dishwashers, we account for the increase in faucet usage associated
with handwashing of dishes. We assume that the amount of hot water used for the
dishwasher is also used for handwashing, with the added cold water used to bring the
temperature of this additional water to 105 degrees F for both flow and filling components
of hand dishwashing. This additional water use is split 50:50 between filling and flow for
Pre-1994 households. For Post-1993 households, the flow part of the additional use is
reduced to reflect the 1994 EPACT flow standards. Table 4 shows the results of the
calculations, weighted by the fraction of homes with and without dishwashers.

3
Table 2: Water temperature assumptions and implied
hot/cold water use breakdowns

Usage
temperature
degrees F % hot % cold
Vintage
Sink filling 105 57% 43%
Faucet flow 80 25% 75%
Bath filling 100 51% 49%
Showers 105 57% 43%
Clotheswasher Existing 83 29% 71%
Post-1993 w/standards 78 22% 78%
Dishwasher 139 100% 0%
Toilets 60 0% 100%
Yard watering/other 60 0% 100%

Hot water temperature = 139 degrees F
Cold water temperature = 60 degrees F
Warm water temperature = 99

(1) New clothes washers assumed to eliminate warm rinse option. Temperature for
clothes washers is a weighted average based on P&G data, as cited in DOE 1990 p. 3-11.
(2) Temperature for dishwashers set at highest allowable level.
(3) Faucet flow assumed to be 50% cold and 50% warm (warm = 50% cold, 50% hot).
(4) Hot water temperature is a weighted average from a survey cited in Bancroft et al. (1991).
(5) Cold water temperature represents a rough annual US average adopted from data in Labs (1979).
(6) Warm water temperature assumed to be a simple average of cold and hot water temperatures.
(7) Shower temperature taken from surveys cited in Brown and Caldwell 1984.
Table 3: Usage times and flow rates for showers, baths, sinks, and faucets.

Flow uses Baths/sink filling
Usage/person Usage/HH Flow rate Total flow Hot flow Cold flow Total Hot filling Cold filling
min/day min/day GPM gals/day gals/day gals/day gals/day gals/day gals/day

Showers/baths Pre-1994 5 13 3.4 45 26 19 19 9.5 9.2
Post-1993 5 13 2.5 33 19 14 19 9.5 9.2

Bathroom faucets Pre-1994 1.2 3.4 4.1 1.0 3.1 4.1 2.3 1.7
Post-1993 1.2 2.5 3.0 0.8 2.3 4.1 2.3 1.7

Kitchen faucets w/DW Pre-1994 1.5 3.4 5.1 1.3 3.8 5.1 2.9 2.2
Post-1993 1.5 2.5 3.8 0.9 2.8 5.1 2.9 2.2

Laundry faucets Pre-1994 0.3 3.4 1.0 0.3 0.8 1.0 0.6 0.4
Post-1993 0.3 2.5 0.8 0.2 0.6 1.0 0.6 0.4

All Faucets w/DW Pre-1994 3.0 3.4 10.2 2.6 7.7 10.2 5.8 4.4
Post-1993 3.0 2.5 7.5 1.9 5.6 10.2 5.8 4.4

Kitchen faucets w/o DW Pre-1994 3.4 3.4 11.6 5.0 6.6 11.6 6.6 5.0
Post-1993 3.4 2.5 8.5 3.7 4.9 11.6 6.6 5.0

All Faucets w/o DW Pre-1994 4.9 3.4 17 6.3 10.4 17 9.6 7.2
Post-1993 4.9 2.5 12.3 4.6 7.7 17 9.6 7.2

(1) All hot and cold water breakdowns taken from Table 2.
(2) Base shower flow rates (3.4 gpm) from Brown and Caldwell 1984. EPACT standard mandates flow rates of 2.5 gpm starting 1 January 1994.
Shower usage of 5 minutes/capita/day adapted from Brown and Caldwell 1984 (1990 US average is 2.67 persons/household).
(3) Total water use for baths using assumptions for per capita use (0.14 baths per capita per day and 50 gals/bath) from Brown and Caldwell 1984,
multiplied by 2.67 persons per household.
(4) Base case faucet flow rates assumed to be the same as shower flow rates (3.4 gpm). EPACT standard mandates flow rates of 2.5 gpm starting 1 January 1994.
Faucet flow usage assumed to be 3 minutes per day per household, with 40% going to bathroom faucets, 50% to kitchen, and 10% to laundry.
(4) Faucet filling is always assumed to be equal to base case flow for bathrooms, kitchens, and laundry faucets, respectively. Filling is not affected by standards.
(5) For homes w/o DW in the "Pre-1994" case, additional hot water use for washing dishes is assumed to equal 7.47 gals/day, which is equal to the usage
of pre-1994 DWs. The associated cold water usage is calculated assuming a delivered water temperature of 105 deg F for the additional water for hand dishwashing.
Additional water use associated with hand dishwashing is split 50:50 between filling and flow. In the "Post-1993" case, flow is reduced by the ratio 2.5gpm/3.4gpm.
2.4 Toilets

Table 4 also shows usage per household for toilets. This usage is based on an assumption
of 4 flushes per person per day (Brown and Caldwell 1984) and 2.67 persons per
household. In the base case, toilets use 4.3 gallons per flush. This value represents the
weighted average for 1983 from Brown and Caldwell (1984), adjusted assuming that all
new toilets installed between 1983 and 1993 use 3.5 gallons per flush. The EPACT
standards reduce this for new toilets after January 1, 1994 to 1.6 gallons per flush.

2.5 Dishwashers and clotheswashers

Dishwasher and clothes washer water use per cycle are taken from US DOE (1990). Total
cycles are taken from a Proctor and Gamble survey as cited in US DOE (1990).
Dishwasher standards save only hot water (because dishwashers only use hot water) while
clothes washer standards save both hot and cold water.

Table 4 shows the results of these calculations for pre-and post-1994 standard washers and
dishwashers on a saturation-weighted basis (all other end-uses [e.g., toilets] are assumed to
have a saturation of 100%). Saturations for dishwashers, clothes washers, and water
heaters are taken from the electronic version of US DOE (1992) Weighted average water
use per household associated with clothes washers and dishwashers is higher for Post-
1993 households than for Pre-1994 homes because of the greater saturations of these
appliances in new houses.

2.6 Other uses

To estimate the size of the "Yard/other" water use, we subtracted the water usage per
household for pre-1994 households for all identified end-uses in Table 4 from the US
Geological Service's estimate of total residential water use per household (Solley et al.
1993) . This approach assumes that the total water usage per household does not change
between 1990 and 1993. We also assume that this residual usage is the same across all
house vintages, appliance vintages, and fixture/fitting vintages, and is constant over the
forecast period.

3. FORECASTING METHODOLOGY AND RESULTS

We create a simple frozen efficiency forecast for both water use and energy use for water
heating. We assume that the saturation and efficiency of equipment existing in the base
year remain the same until that equipment is replaced. The saturation and efficiency of new
equipment installed throughout the forecast remain at the same level as saturations and
efficiencies for new devices in the base year. Equipment retires at an exponential rate equal
to the inverse of the equipment lifetime.

3.1 Forecasted number of households

Table 5 shows the forecast of number of households, taken from US DOE (1994). We
disaggregated this forecast into 4 different combinations of house vintages, appliance
vintages, and fixture/fitting vintages. We use exponential retirement rates from Koomey et
al. (1993) as stated in the Table to estimate for any year the remaining stock of houses
associated with each combination of equipment vintages.

6
Table 4: Saturation-weighted hot and cold water usage per household by end use, house vintage, and appliance vintage

Fixtures and
Home Appliance fittings Sink filling Faucet flow Bath filling Showers Clotheswasher Dishwasher Toilets Yard/other Total
vintage vintage vintage gals/day gals/day gals/day gals/day gals/day gals/day gals/day gals/day gals/day
Baseline
Hot water Pre-1994 All All 7.9 4.6 9.5 26.0 8.2 3.4 0.0 0.0 59.5
Post-1993 All All 6.7 3.4 9.5 26.0 9.8 5.8 0.0 0.0 61.1

Cold water Pre-1994 All All 5.9 9.2 9.2 19.4 20.3 0.0 59.3 86.4 209.7
Post-1993 All All 5.0 8.3 9.2 19.4 24.2 0.0 59.3 86.4 211.8

Standards case
Hot water Pre-1994 Post-1993 Post-1993 7.9 3.4 9.5 19.1 5.6 2.7 0.0 0.0 48.1
Pre-1994 Post-1993 Pre-1994 7.9 4.6 9.5 26.0 5.6 2.7 0.0 0.0 56.2
Pre-1994 Pre-1994 Pre-1994 7.9 4.6 9.5 26.0 8.2 3.4 0.0 0.0 59.5
Post-1993 Post-1993 Post-1993 6.7 2.5 9.5 19.1 6.6 4.6 0.0 0.0 49.0

Cold water Pre-1994 Post-1993 Post-1993 5.9 6.7 9.2 14.3 19.4 0.0 29.9 86.4 171.9
Pre-1994 Post-1993 Pre-1994 5.9 9.2 9.2 19.4 19.4 0.0 59.3 86.4 208.8
Pre-1994 Pre-1994 Pre-1994 5.9 9.2 9.2 19.4 20.3 0.0 59.3 86.4 209.7
Post-1993 Post-1993 Post-1993 5.0 6.1 9.2 14.3 23.2 0.0 29.9 86.4 174.1

(1) Saturation of CW is 76% and 91% for Pre-1994 and Post-1993 homes, respectively. Saturation of DW is 45% and 77% for Pre-1994 and Post-1993
homes, respectively. Source: RECS 1990; Post-1993 saturations derived by calculating saturations for homes built 1985-90 (US DOE 1992).
All dishwasher owners are assumed to own clotheswashers. Existing homes without DW or CW do not acquire them during the analysis period.
Appliance saturations for Post-1993 homes also remain constant over the analysis period. Saturations of sink filling, faucets, showers, baths, toilets, and other is 100%.
(2) Reference case is frozen efficiency: existing equipment remains at 1990 average efficiency until replaced by new devices. New devices always
have same efficiency as new devices in 1990.
(3) Faucet and shower usage from Table 3.
(4) Hot water use per appliance for pre-1994 CW is 10.72 gals/day; for post-1993 CW is 7.29 gals/day. Hot water use per appliance for pre-1994 DW
is 7.47 gals/day; for post-1993 DW is 5.96 gals/day. Source: US DOE (1990). Usage estimates (CW cycles per year = 380. DW cycles per year = 229)
taken from a Proctor and Gamble survey cited in US DOE (1990).
(5) Toilets flushed 4 times per person per day (2.67 persons/household) from Brown and Caldwell 1984. Capacity of average existing toilet is 4.3 gals/flush,
while new toilets meeting EPACT standards use 1.6 gals/flush. Toilet water use includes leakage (20% of the toilets leak 24 gals/capita per day; taken from Brown and Caldwell 1984).
(6) Water use for yard and other uses (e.g. car washing, window washing) is calculated as the difference between total US domestic water use per household (from Solley et al 1993)
and the sum of water use per household for kitchens, bathrooms, dishwashers, and clotheswashers from our bottom-up calculations. We assume in this calculation that overall
water use per household does not change between 1990 and 1993.
Table 5: Number of US households by home vintage, appliance vintage, and fixtures/fittings vintage

Fixtures and
Home Appliance fittings Millions of households
vintage vintage vintage 1993 2000 2005 2010

Number of households Pre-1994 Post-1993 Post-1993 0.0 23.9 35.5 43.9
Pre-1994 Post-1993 Pre-1994 0.0 12.2 15.1 15.5
Pre-1994 Pre-1994 Pre-1994 95.9 54.8 36.7 24.6
Post-1993 Post-1993 Post-1993 0.0 10.9 18.5 26.4
All All All 95.9 101.7 105.9 110.5

% of households Pre-1994 Post-1993 Post-1993 0% 23% 34% 40%
Pre-1994 Post-1993 Pre-1994 0% 12% 14% 14%
Pre-1994 Pre-1994 Pre-1994 100% 54% 35% 22%
Post-1993 Post-1993 Post-1993 0% 11% 18% 24%
All All All 100% 100% 100% 100%

(1) Baseline household forecast from US DOE (1994).
(2) 1993 households derived assuming a linear absolute annual increase in households between 1992 and 2000.
(3) Lifetimes for clothes washers, dishwashers, and water heaters assumed to be 13 years, based on Hanford et al. (1994).
(WH and DW life = 13 years, CW life = 14 years. We assume 13 years for all to reduce complexity).
Lifetimes for fittings and fixtures assumed to be 20 years.
(4) Appliances retire faster than fittings and fixtures, so we assume for simplicity that there are no homes with
new faucets/SHs and old appliances.
(5) Home retirement rates taken from Koomey et al. (1993) are 0.77%, averaged over all house types.
3.2 Baseline and standards case forecasts: water use

To calculate total water use, we multiply the water usage per household in each vintage bin
in a given year by the number of households in each bin. Table 6 shows the resulting hot
and cold water usage by end-use for the US as a whole for 1993 and 2010. The largest
end-use by far is "Yard/other" which represents about one-third of total water use in the
base case. Next is toilets, followed by showers and clothes washers. Showers, bath
filling, and clothes washers are the largest hot water end-uses. Total water savings is
largest for toilets, followed by showers and clothes washers.

3.3 Baseline and standards case forecasts: energy use

We use the saturations of different water heater fuel types (new and existing), the hot water
usage in Table 6, the standby losses and recovery efficiencies from Table 1, and the
equations above to calculate total energy use in the baseline and standards cases.
Saturations for water heaters are taken from the electronic version of US DOE (1992)
(saturations for new equipment are derived using the saturation of water heaters for homes
built 1985-1990). Table 7 shows the results of these calculations. For electric water
heating, showers are the largest end-use, followed by standby losses, bath filling, and
clothes washers. For gas and oil water heating, standby losses are the largest end-use
followed by showers, bath filling, and clothes washers.

3.4 Energy savings

Table 8 shows the savings from the standards by end-use. The 1990 water heater
standards only affect standby losses, while the 1994 fixture/fitting, clothes washer, and
dishwasher standards affect only flow rates for these end-uses.

For electric water heaters, the flow-related standards save more than twice as much
electricity as the 1990 water heater standards. For gas and oil-fired water heaters, the 1990
water heater standards save about 60% more energy than the flow related standards. Of the
flow related standards, showerhead standards save the most, followed by clothes washers
and dishwashers. Total savings in site energy terms are about 0.5 quads per year in 2010,
while savings in primary energy terms would be about 0.8 quads per year in 2010
(assuming electricity is valued at 11,157 Btus/kWh).

4. FUTURE WORK

4.1 Sensitivity analysis/further disaggregation

This framework would easily allow testing of the sensitivity of the forecast results to
changes in different parameters. Water usage varies by house type, number of persons per
household, age of occupants, water prices, water rate structures, source of water (public
systems or wells), geography, and income levels. In particular, the effect of standards on
households with different appliance ownership and water usage patterns could easily be
tested within this framework. An examination of extreme cases (exceptionally low or high
water use) would also be illuminating. We believe that creating such a sensitivity analysis
should await the arrival of more accurate data on water usage patterns and how these
patterns vary across the country.

9
Table 6: U.S. hot and cold water usage and savings by end use

Billion gallons per day
Year Sink filling Faucet flow Bath filling Showers Clotheswasher Dishwasher Toilets Yard/other Total

Baseline
Hot water 1993 0.75 0.44 0.91 2.49 0.78 0.33 0.00 0.00 5.7
2010 0.84 0.48 1.05 2.87 0.95 0.44 0.00 0.00 6.6
Cold water 1993 0.57 0.88 0.88 1.86 1.94 0.00 5.68 8.29 20.1
2010 0.63 0.99 1.02 2.15 2.34 0.00 6.55 9.55 23.2
Total water 1993 1.32 1.32 1.79 4.35 2.73 0.33 5.68 8.29 25.8
2010 1.47 1.47 2.06 5.01 3.29 0.44 6.55 9.55 29.8

Standards
Hot water 1993 0.75 0.44 0.91 2.49 0.78 0.33 0.00 0.00 5.7
2010 0.84 0.40 1.05 2.38 0.71 0.37 0.00 0.00 5.7
Cold water 1993 0.57 0.88 0.88 1.86 1.94 0.00 5.68 8.29 20.1
2010 0.63 0.83 1.02 1.78 2.27 0.00 4.48 9.55 20.5
Total water 1993 1.32 1.32 1.79 4.35 2.73 0.33 5.68 8.29 25.8
2010 1.47 1.22 2.06 4.17 2.97 0.37 4.48 9.55 26.3

Savings
Hot water 1993 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0
2010 0.00 0.08 0.00 0.48 0.24 0.07 0.00 0.00 0.9
Cold water 1993 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0
2010 0.00 0.16 0.00 0.36 0.08 0.00 2.07 0.00 2.7
Total water 1993 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0
2010 0.00 0.24 0.00 0.85 0.32 0.07 2.07 0.00 3.5

(1) Total usage calculated using weighted average daily per household usage from Table 4 and housing populations from Table 5.
(2) Increase in cold water usage for clotheswasher in standards case caused by elimination of warm rinse cycles and increasing saturation of clotheswashers in new buildings.
(3) Toilet water use includes leakage.
Table 7: Forecasted US energy use for water heaters by end-use in the baseline and standards cases

Use-related water heating energy consumption
Year Fuel type Sink filling Faucet flow Bath filling Showers Clotheswasher Dishwasher Standby losses Total
TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr

Frozen efficiency baseline 1993 Electric resistance 69 40 83 227 72 30 107 628
Natural gas 135 79 163 445 140 58 993 2012
Oil 13 7 16 42 13 6 95 192

2010 Electric resistance 87 49 110 301 101 49 129 825
Natural gas 139 79 172 470 153 69 1015 2096
Oil 12 7 15 40 13 6 87 179

Baseline + 1990 WH stds 1993 Electric resistance 69 40 83 227 72 30 98 618
Natural gas 135 79 163 445 140 58 916 1935
Oil 13 7 16 42 13 6 87 185

2010 Electric resistance 87 49 110 301 101 49 92 788
Natural gas 139 79 172 470 153 69 802 1883
Oil 12 7 15 40 13 6 69 161

Baseline + 1990 WH stds + 1993 Electric resistance 69 40 83 227 72 30 98 618
1994 fixture/fitting/CW/DW Natural gas 135 79 163 445 140 58 916 1935
standards Oil 13 7 16 42 13 6 87 185

2010 Electric resistance 87 40 110 246 75 40 92 690
Natural gas 139 67 172 395 115 58 802 1747
Oil 12 6 15 34 10 5 69 151

(1) Energy use for clotheswashers and dishwashers does not include motor energy, only the energy needed to heat water.
(2) Electricity measured in site energy at 3412 Btu/kWh.
(3) Saturations of water heating fuel assumed to remain constant throughout the analysis period for existing homes (no conversions).
New homes throughout the analysis period are assumed to have the same saturation as new homes in 1990.
(4) WH = water heating; CW = clothes washer; DW = dishwasher.
(5) Saturations of electric water heaters: Pre 1994=38%, Post 1993=61%; Gas WHs: Pre 1994=57%, Post 1993=37%; Oil WHs: Pre 1994=5%, Post 1993=1%.
Source of saturations: RECS 1990; Post-1993 saturations derived by calculating saturations for homes built 1985-90 (US DOE 1992).
Table 8: Forecasted US energy savings by end-use for water heaters in the standards cases in 2010

Use-related water heating energy use
Units Fuel type Sink filling Faucet flow Bath filling Showers Clotheswasher Dishwasher Standby losses Total
TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr TBtu/yr

1990 WH stds TBtu/yr Electric resistance 0 0 0 0 0 0 37 37
Natural gas 0 0 0 0 0 0 214 214
Oil 0 0 0 0 0 0 18 18

% of total Electric resistance 0% 0% 0% 0% 0% 0% 100% 100%
Natural gas 0% 0% 0% 0% 0% 0% 100% 100%
Oil 0% 0% 0% 0% 0% 0% 100% 100%

1994 fixture/fitting/CW/DW standards TBtu/yr Electric resistance 0 8 0 55 27 8 0 98
Natural gas 0 12 0 75 37 11 0 136
Oil 0 1 0 6 3 1 0 11

% of total Electric resistance 0% 9% 0% 56% 27% 8% 0% 100%
Natural gas 0% 9% 0% 55% 28% 8% 0% 100%
Oil 0% 9% 0% 55% 28% 8% 0% 100%

1990 WH stds + 1994 fixture/ TBtu/yr Electric resistance 0 8 0 55 27 8 37 135
fitting/CW/DW standards Natural gas 0 12 0 75 37 11 214 349
Oil 0 1 0 6 3 1 18 29

% of total Electric resistance 0% 6% 0% 41% 20% 6% 27% 100%
Natural gas 0% 4% 0% 21% 11% 3% 61% 100%
Oil 0% 4% 0% 21% 11% 3% 62% 100%

We have not accounted for energy losses associated with transporting hot water from the
water heater to the tap. It is not certain how large these losses are, but they may be
significant. Multi-family buildings with recirculating water systems are totally different
than single family dwellings in this regard, and probably should be analyzed separately in
the future. More measurements of such effects are needed.

4.3 Measurement of water use

It is crucially important that future programs to measure water use take a comprehensive
bottom-up perspective like the one in this paper. Measurements of hot water consumption
alone are not sufficient. Cold water use, water heater set point, cold water inlet
temperature, and usage temperature must also be measured. In addition, the results must
be correlated with detailed information on the characteristics of each household (including
saturations and efficiencies of water-using devices) because estimates of water use in the
aggregate have little meaning. Finally, accurate measurements of actual usage behavior
(i.e., number of flushes per day, minutes of faucet on-time) are also crucial to estimating
total water use and the potential for water savings.

4.4 Comparison to other forecasts

There are several other widely used forecasts of water heater energy use to which the
estimates developed here could be compared (Geller and Nadel 1992, Koomey et al. 1993,
US DOE 1994). We have not undertaken this comparison for the sake of brevity, but any
further work using this framework should do so. The documentation for these forecasts
typically does not include the detailed information needed to make such a comparison, but
this information can often be inferred from conversations with the modelers.

4.5 "Upstream" and "Downstream" issues

We have ignored energy, economic, and technological issues related to pumping and
treatment of water. These issues are complex and have not typically been included in
policy studies that focus on direct energy savings from more efficient end-using
equipment. They may in some cases be important, especially when conducting cost-
benefit analyses in regions where the choice for the local water district may be efficiency
policies or adding peak capacity to its water treatment facilities. These issues can also be
important for those users not connected to the public water system, because well pumping
and septic tanks are substantially affected by changes in total water usage.

4.6 Enforcement of standards

If low-flow devices do not deliver equivalent service, then customer satisfaction may lead
to circumvention of regulations or changes in usage patterns (i.e., longer showers). We
have assumed in this analysis that the low flow devices work exactly as advertised 100%
of the time, and that they do not affect usage behavior. To improve the accuracy of the
forecasts, we need data on whether this assumption is even roughly correct.

5. CONCLUSIONS

13
This paper describes a detailed end-use breakdown of hot and cold water use, and uses this
breakdown to estimate the effects of recent appliance and fixture/fitting standards on water
use and energy use. Such a detailed and comprehensive breakdown is essential for truly
understanding the water heating end-use. Previous efforts at measurement have generally
been inadequate or incomplete, and we hope that this attempt to systematize the various
data will serve as a model for future measurements and analysis.

Forecasted water savings from the EPACT standards are largest for toilets, followed by
showers, clothes washers, and dishwashers. Hot water savings are greatest for showers
and clothes washers, followed by faucets and dishwashers. Forecasted standby loss
energy savings from the 1990 electric water heater standards are only about one third as
large as electricity savings associated with the EPACT 1994 flow standards. For natural
gas water heaters the situation is reversed, with standby loss energy savings from the 1990
gas water heater standards being about 60% larger in magnitude than gas savings
associated with the EPACT 1994 flow standards. Overall primary energy savings
associated with current standards affecting water heaters and hot water use are projected to
be about 0.8 quadrillion Btus/year by 2010.

ACKNOWLEDGEMENTS

We are grateful to Mark Friedrichs of the US Department of Energy for supporting this
work. We are also grateful for the careful reviews of Richard Brown (LBL), Celina
Atkinson (LBL), Mary Suttor (Quantum Consulting), and Frances Wood (AES). The
work described in this paper was supported by the Office of Energy Efficiency and
Alternative Fuels Policy, Office of Policy of the U.S. Department of Energy under
Contract Number DE-AC03-76SF00098.

REFERENCES

Bancroft, Brady, Michael Shepard, Amory B. Lovins, Robert C. Bishop, Henry G. Patton,
Linda Baynham, John Wilson, and James Johnson. 1991. The State of the Art:
Water Heating. Competitek. October.

Brown and Caldwell. 1984. Residential Water Conservation Projects: Summary Report.
US Department of Housing and Urban Development. June.

Geller, Howard S., and Steven M. Nadel. 1992. "Consensus National Efficiency Standards
for Lamps, Motors, Showerheads and Faucets, and Commercial HVAC
Equipment". Presented at ACEEE 1992 Summer Study on Energy Efficiency in
Buildings in Asilomar, CA. Published by American Council for an Energy
Efficiency Economy.

Hanford, James W., Jonathan G. Koomey, Lisa E. Stewart, Matthew E. Lecar, Francis X.
Johnson, and Roland J. Hwang. 1994. Baseline Data for the Residential Sector
and Development of a Residential Forecasting Database. Lawrence Berkeley
Laboratory. LBL-33717. April.

Kempton, Willet. 1984. "Residential Hot Water: A Behaviorally-Driven System." In
Proceedings of the 1984 ACEEE Summer Study on Energy Efficiency in Buildings.
Santa Cruz, CA. American Council for an Energy Efficient Economy.

14
Koomey, Jonathan, Francis X. Johnson, James E. McMahon, Mary Orland, Mark Levine,
Peter Chan, and Florentin Krause. 1993. An Assessment of Future Energy Use and
Carbon Emissions from U.S. Residences. Lawrence Berkeley Laboratory. LBL-
32183. December.

Labs, Kenneth. 1979. "Underground Building Climate". Solar Age. October. p. 44.

Perlman, Maier. 1987. "Residential Water Heating: Low-Tech and High-Tech
Alternatives". Energy Auditor and Retrofitter. January/February. p. 25.

Rohles Jr., Frederick H. , and Stephan A. Konz. 1981. Showering Behavior: Implications
for Water and Energy Conservation. Institute for Environmental Research, Kansas
State University. 81-02. April.

Solley, Wayne B., Robert R. Pierce, and Howard A. Perlman. 1993. Estimated Use of
Water in the United States in 1990. US Geological Service (USGS). Circular
1081.

Taylor, M. E., K.G. Ritland, and R. G. Pratt. 1991. Hot Water Electric Energy Use in
Single-Family Residences in the Pacific Northwest: Regional End-Use Metering
Project. Office of Energy Resources, Bonneville Power Administration.
DOE/BP-13795-27. September.

Thrasher, William H., D. W. DeWerth, and B. R. Becker. 1990. Comparison of Collected
and Compiled Existing Data on Service Hot Water Use Patterns in Residential and
Commercial Establishments. ASHRAE. Research Project No. 600-RP. August 8.

US DOE, U.S. Department of Energy. 1990. Technical Support Document: Energy
Conservation Standards for Consumer Products: Dishwashers, Clothes Washers,
and Clothes Dryers. U.S. Department of Energy, Assistant Secretary,
Conservation and Renewable Energy, Building Equipment Division. DOE/CE-
0299P. December.

US DOE, U.S. Department of Energy. 1992. Residential Energy Consumption Survey:
Housing Characteristics 1990. EIA, Energy Information Administration.
DOE/EIA-0314(90). May.

US DOE, U.S. Department of Energy. 1993. Technical Support Document: Energy
Conservation Standards for Consumer Products: Room Air Conditioners, Water
Heaters, Direct Heating Equipment, Mobile Home Furnaces, Kitchen Ranges and
Ovens, Pool Heaters, Fluorescent Lamp Ballasts, and Television Sets. U.S.
Department of Energy, Assistant Secretary, Energy Efficiency and Renewable
Energy, Building Equipment Division. Volume 3: Water Heaters, Pool Heaters,
Direct Heating Equipment, Mobile Home Furnaces; DOE/EE-0009. November.

US DOE, U.S. Department of Energy. 1994. Annual Energy Outlook, with Projections to
2010. Energy Information Administration. DOE/EIA-0383(94). January.

Warwick, W. M. 1993. Energy Savings from Energy-Efficient Showerheads: REMP Case
Study Results, Proposed Evaluation Algorithm, and Program Design
Implications. Prepared for the Bonneville Power Administration by Pacific
Northwest Laboratory. September.