ArticlePDF Available

Case study field evaluation of a systems approach to retrofitting a residential HVAC system

Authors:

Abstract and Figures

This case study focusing on a residence in northern California was undertaken as a demonstration of the potential of a systems approach to HVAC retrofits. The systems approach means that other retrofits that can affect the HVAC system are also considered. For example, added building envelope insulation reduces building loads so that smaller capacity HVAC system can be used. Secondly, we wanted to examine the practical issues and interactions with contractors and code officials required to accomplish the systems approach because it represents a departure from current practice. We identified problems in the processes of communication and installation of the retrofit that led to compromises in the final energy efficiency of the HVAC system. These issues must be overcome in order for HVAC retrofits to deliver the increased performance that they promise. The experience gained in this case study was used to optimize best practices guidelines for contractors (Walker 2003) that include building diagnostics and checklists as tools to assist in ensuring the energy efficiency of ''house as a system'' HVAC retrofits. The best practices guidelines proved to be an excellent tool for evaluating the eight existing homes in this study, and we received positive feedback from many potential users who reviewed and used them. In addition, we were able to substantially improve the energy efficiency of the retrofitted case study house by adding envelope insulation, a more efficient furnace and air conditioner, an economizer and by reducing duct leakage.
Content may be subject to copyright.
eScholarship provides open access, scholarly publishing
services to the University of California and delivers a dynamic
research platform to scholars worldwide.
Lawrence Berkeley National Laboratory
Lawrence Berkeley National Laboratory
Title:
Case study field evaluation of a systems approach to retrofitting a residential HVAC system
Author:
Walker, Iain S.
McWiliams, Jennifer A.
Konopacki, Steven J.
Publication Date:
09-01-2003
Publication Info:
Lawrence Berkeley National Laboratory
Permalink:
http://escholarship.org/uc/item/0tz5v4st
DRAFT – DO NOT QUOTE
LBNL-53444
Case Study Field Evaluation of a Systems Approach to
Retrofitting a Residential HVAC System
Iain Walker, Jennifer McWilliams and Steven Konopacki
Energy Performance of Buildings Group
Lawrence Berkeley National Laboratory
Berkeley, California 94720
September 2003
This work was supported by the Assistant Secretary for Energy Efficiency and Renewable
Energy, Building Technologies Program, U. S. Department of Energy (DOE) under contract No.
DE-AC03-76SF00098.
DRAFT – DO NOT QUOTE
Abstract
This case study focusing on a residence in northern California was undertaken as a
demonstration of the potential of a systems approach to HVAC retrofits. The systems approach
means that other retrofits that can affect the HVAC system are also considered. For example,
added building envelope insulation reduces building loads so that smaller capacity HVAC
system can be used. Secondly, we wanted to examine the practical issues and interactions with
contractors and code officials required to accomplish the systems approach because it represents
a departure from current practice. We identified problems in the processes of communication
and installation of the retrofit that led to compromises in the final energy efficiency of the
HVAC system. These issues must be overcome in order for HVAC retrofits to deliver the
increased performance that they promise. The experience gained in this case study was used to
optimize best practices guidelines for contractors (Walker 2003) that include building
diagnostics and checklists as tools to assist in ensuring the energy efficiency of “house as a
system” HVAC retrofits. The best practices guidelines proved to be an excellent tool for
evaluating the eight existing homes in this study, and we received positive feedback from many
potential users who reviewed and used them. In addition, we were able to substantially improve
the energy efficiency of the retrofitted case study house by adding envelope insulation, a more
efficient furnace and air conditioner, an economizer and by reducing duct leakage.
DRAFT – DO NOT QUOTE
Table of Contents
Introduction ............................................................................................................................... 4
Diagnostics and Screening (D&S) Process .............................................................................. 5
Diagnostics and Screening Results from Four Houses......................................................... 14
Retrofit Selection ..................................................................................................................... 24
Diagnostics and Screening in Four Cold Climate Houses ................................................... 38
Lessons Learned for Best Practices Guidelines .................................................................... 39
References ................................................................................................................................ 42
Appendix A. Field Surveys of Four Houses in California for Retrofitting ....................... 45
Appendix B: Results ............................................................................................................... 49
Appendix C. Calulation of Operating Energy Efficiency Ratio (EER) and Cooling
Capacity.................................................................................................................................... 62
Appendix D. Residential Commissioning Procedures......................................................... 63
Appendix E: Schematic diagrams of the HVAC system...................................................... 69
Appendix F. Retrofit material and labor costs billed by the HVAC contractor. ............. 71
Appendix G. Field Surveys of Four Cold Climate Houses for Retrofitting...................... 74
3
DRAFT – DO NOT QUOTE
Introduction
The HVAC equipment installed in an energy efficient retrofit operates at the rated high
efficiency only when the contractor performs an installation that adheres to strict standards. In
this case study of a contracted residential HVAC energy efficient retrofit we describe our
experiences of retrofitting a house with two contractors. We also describe the role of the local
building code official. The retrofit utilizes a systems approach, which integrates many aspects of
the building and Heating Ventilating and Air-Conditioning (HVAC) system to achieve improved
performance compared to replacing individual components independently. This system-wide
retrofit focuses on four areas of the building system: [1] building envelope sealing [2] additional
attic floor insulation [3] duct sealing and additional duct insulation [4] down-sized high-
efficiency heating and cooling equipment. Because of the dry summer climate the retrofit
included an economizer. Based on occupant complaints of severe stratification we added a zone
control to allow better thermal control of the two floors of the house.
Rather than simply replace an HVAC system with equipment of the same energy efficiency,
retrofits are an opportunity to use higher efficiency equipment. The HVAC system can have
many added features that ensure increased comfort, safety and durability in addition to reduced
energy use. Examples include: [A] multiple-speed heating or cooling equipment to better match
building loads [B] added economizers to provide ventilation and reduce electricity consumption,
and [C] added zoning to increase comfort - this is particularly useful in houses that have large
areas that are poorly conditioned. A classic example is the two-story house that currently
operates as a single zone and does not provide enough cooling upstairs – a very common
complaint.
The methodology to conduct the case study consisted of seven tasks: [1] pre-retrofit diagnostics
to screen the building and HVAC systems of several homes to select a residence for the study [2]
monitor the test residence over several months before and after the retrofit for energy and
comfort performance [3] perform system-wide retrofits [4] post-retrofit diagnostics of the
building and HVAC systems to measure improvement from retrofits [5] identify problems in the
processes of communication and installation of the retrofit and evaluate quality of workmanship
[6] propose solutions such that the HVAC equipment can provide energy efficient and
comfortable cooling and heating for the occupants and apply these lessons to the best practices
guidelines for contractors [7] develop guidelines and protocols for contractors.
To facilitate finding a house suitable for retrofit and obtaining timely homeowner permission to
retrofit and monitor the house, we used a house volunteered and occupied by Lawrence Berkeley
National Laboratory (LBNL) staff (homeowner acceptance of these types of retrofits and
monitoring projects can be a barrier to obtaining access for use of a house). The retrofit
combined readily available residential HVAC products into a system that can be regarded as an
example and a basis for future work. The selection of components (e.g., equipment sizing) was
based on an engineering analysis performed by LBNL combined with our previous experience in
the field monitoring the performance of thermal distribution systems (Walker et al. 1999, Walker
and Modera 1998, Walker et al 1998, Jump et al 1996, Jump and Modera 1994).
4
DRAFT – DO NOT QUOTE
Diagnostics and Screening (D&S) Process
A key aspect in the design of retrofits is the need to know the current performance of a house in
order to understand which building components have the potential for improvement. To perform
this task LBNL compiled a set of diagnostic screening tools that combine physical
measurements, observations and a homeowner questionnaire. The measured test results,
observations and homeowner answers to questions are used to direct the HVAC contractor or
designer towards the best retrofits applicable to each individual house. The retrofits will depend
on the current condition of the building envelope and HVAC system, the local climate, the
construction methods used for the house, the presence of various energy reduction systems
and/or materials, and issues that would allow particular energy saving systems (e.g., other
retrofits or alterations to the house being performed at the same time as the HVAC retrofit). This
approach is similar to a doctor referring a patient for blood tests or x-rays before actually
performing surgery, where the doctor can diagnose the patient and be confident that he/she does
the right thing. To take this analogy further – we can borrow from the medical profession and
say that the first thought when retrofitting a house is to do no harm, i.e., do not make changes
that could make the house worse to live in.
The checklist is not definitive and we expect that items will be removed, changed or added over
time. For example, in our field testing we found that it was time consuming and requires
someone with electrical training to determine the power consumption of all the individual
elements of the HVAC system, so it is unlikely that this diagnostic would be part of an HVAC
contractors toolkit – although for research level activities these tests provide vital information on
system performance. Another example is the addition of pressure drop measurements across
filters, where we found in one of our test houses that the homeowner had replaced the original
poor filters with high efficiency MERV 12 filters. This change introduced considerable
additional pressure drop in the system that the system had not been designed for. In this case we
would recommend a retrofit that reduced flow resistance in the system (e.g., by replacing high
resistance ducts or by adding return ducts) or replace the filters and filter rack with deeper
element filters (i.e., 2 inch rather than one inch filters). These deeper filters offer the same
filtration efficiency with increased surface area and therefore lower flow resistance.
The diagnostic screening tools are in checklist format and consist of three parts: [1] HVAC [2]
envelope [3] and occupant survey. A template of the screening checklist is displayed in Table 1.
This table includes a column for the actual value found during the D & S process. The results of
the diagnostics are discussed in the Diagnostics and Screening Results chapter. Detailed
protocols for performing the diagnostics are presented in Appendix D.
5
DRAFT – DO NOT QUOTE
Table 1. Diagnostics Screening Checklist: HVAC, Envelope and Occupant Survey.
Measurement/Obser
vation
Potential Target value Actual
Value
Potential Retrofit Action
Duct leakage <10% of air handler flow Seal ducts: Aeroseal/tape/mastic
Duct insulation R6 (RSI 1) to R8 (RSI 1.4) for all ducts
outside conditioned space
Add insulation to ducts
Air flows at registers Compare to ACCA manual J Replace registers, open/close dampers, reduce system
flow resistance by straightening existing ducts or
replacing them with straight runs of new ducts.
Air handler flow
Cooling: >400 cfm/ton in dry climate,
or >350 cfm/ton in humid climate
Heating: 12.5 cfm/kBtu/h
Replace filters, fix duct restrictions, change fan speed,
replace fan with high efficient unit, add extra returns in
return restricted systems
Filter Condition Clean and at least MERV 6
1
Replace with MERV 6 or better. Use 2 or 4 inch filters
if possible
Thermostat Setting
Heating: 68°F (20°C) Cooling: 78°F
(25°C)
Thermostat raised in summer and lowered in winter to
account for better distribution, mixing and envelope
improvements.
Spot ventilation 50 cfm each bathroom
100 cfm each kitchen
Replace fans, fix restrictive ducting
Spot Ventilation fan power
consumption
2.5 cfm/W (1.2 L/s/W). A good source
for these ratings is the HVI directory
(www.hvi.org
)
Replace with higher efficiency unit, remove/reduce duct
flow restrictions, clean fan and ducting
Equipment capacity ACCA Manual S Replace with correct size
Refrigerant charge Use superheat or subcooling tests Add/subtract refrigerant
Age and Condition of HVAC
system
Clean and undamaged.
Determine system age.
Clean the system and repair damage or Replace the
system if > 15 years old
Location of HVAC system
equipment and ducts
Inside conditioned space
Seal and insulates duct locations to make them more like
conditioned space, or move system location.
Window A/C units
EnergyStar compliant
Replace with central unit or improved distribution
Multiple systems/zoning System and controls in good working
order and providing good comfort for
occupants
Ensure correct damper operation, check capacity of each
system/zone matches a Manual J (or equivalent) load
calculation
Envelope leakage Normalized Leakage Area reduction of
0.35
Insulate envelope, seal windows/doors/other openings
Moisture testing No moisture problems Source control – better kitchen and bath venting, fix
flashing/detailing, seal and condition crawlspaces in
high humidity climates, replace windows, add insulation
to walls, floors and ceiling
House insulation
Ceiling: R-30 (RSI 5.3) minimum, R-49
(RSI 8.6) in cold/severe cold climate.
Floor over crawlspace:R- 25 (RSI 4.4).
Basement walls: R10 (RSI 1.8), Basement
Floor or slab usually depends on local
codes.
Walls: Cavity should be completely filled
with insulation.
Add insulation to fill cavity. Add semi-permeable rigid
exterior insulation in cold/severe cold climates if the
wall is 2×4 construction.
Windows Double-glazed, low-e. Shaded in
cooling dominant climates
Replace windows. Add shading.
Window shading
Located on south and/or west facing
windows
Add shading to reduce solar loads
Solar radiation control Radiant barrier in attic, low absorptivity
roof coatings
Add radiant barrier in attic, or low absorptivity roof
coatings
Wall, floor and ceiling
construction
Space for ducts/vents
Evaluate house energy bills (if
available)
Occupant survey
Ask occupants to report
problems
No problems Moisture removal strategies, new windows (for
condensation resistance), change register type, airflow
and location to improve mixing/remove drafts, add
envelope insulation, etc.
1
MERV is an industry standard rating system for air filters, it stands for Minimum Efficiency
Report Value determined using ASHRAE Standard 52.2
6
DRAFT – DO NOT QUOTE
Duct Leakage
An ideal duct system would have no duct leakage, however, the components, materials and
installation techniques used for residential systems have led to most duct systems having
significant leakage. Sealing these duct leaks is a key aspect of any HVAC retrofit because the
energy savings can be significant (typically about 25-30%). In addition, the system delivers
more of its conditioned air to the conditioned space instead of losing it to the duct surroundings
(usually attics, crawlspaces, garages or basements), which allows the use of reduced capacity
equipment to meet the same building load. With current residential HVAC duct construction
techniques, materials, and equipment it is very difficult to get to zero duct leakage. For this
reason, recent specifications for duct leakage limits have allowed a small amount of leakage. For
example, the Duct Efficiency Credit in the Alternative Calculations Manual procedure of the
California State Energy Code (Title 24) (http://www.energy.ca.gov/title24/
) allows duct air
leakage at 25 Pa to be 6% of air handler flow for new construction. The US Environmental
Protection Agency EnergyStar ducts page
(http://www.energystar.gov/index.cfm?c=ducts.pr_ducts) sets a limit of 10% for duct leakage
measured using the same technique as Title 24
(http://www.energystar.gov/ia/products/heat_cool/ducts/Duct_Spec_2002.pdf
). An upper limit
of 10% duct leakage is more appropriate when retrofitting existing systems as the 6% level is
often difficult to reach.
There are several methods of duct leakage testing that have been developed for residential
systems. In this study we used the most common method (duct pressurization) and a relatively
new, more sophisticated method called DeltaQ. Both of these methods will be included in future
versions of the ASTM standard for duct leakage E1554.
The Delta Q (described in more detail in Walker et al. 2001 and Walker et al. 2002) test
determines duct leakage flows by measuring the difference in flow through the house envelope
when the house is maintained at a constant envelope pressure differential and the HVAC fan is
turned off and on. The flows are measured at a number of different envelope pressure
differentials, and then used to calculate duct leakage in the supply and return duct systems at
operating conditions. Three major advantages of this test are: [1] it determines the air flow
leakage to outside at operating conditions whereas tests such as duct pressurization determine the
air flow leakage at a given pressure difference which may be quite different from the actual
pressure difference that varies throughout the system [2] it does not require the sometimes
difficult task of covering registers, and [3] in the course of performing this test, all the
measurements are made for a multiple point fan pressurization test to determine envelope
leakage so you get "two tests for the price of one".
The Duct Pressurization test is analogous to the fan pressurization test for envelope leakage. It
measures the leakage through the ductwork at a fixed pressure induced by an external fan. The
procedure involves sealing all the intentional openings in the ductwork (register grilles), and
installing the external fan in an opening to the ductwork (usually in a return grille or at the air
handler cabinet). If the supply and return leakage are desired separately then a blockage is
placed between the supply and return side of the system (often in the filter slot). A single point
or multi point test can be performed in the same manner as an envelope leakage test. Duct
7
DRAFT – DO NOT QUOTE
leakage is often reported in CFM
25
, the flow at 25 Pa pressure differential across the duct
boundary. The 25 Pa reference pressure is typical of an average pressure that the duct boundary
might see in normal operation.
Ducts are often located in spaces that can be considered partly inside and partly outside of the
thermal boundary such as spaces between the floors, or in wall cavities. In these spaces some of
the losses from the ducts will be recovered into the living space through convection or
conduction. In order to determine what fraction of leakage actually leaves the building, there is a
variation of this test where the building envelope is pressurized to 25 Pa at the same time that the
ductwork is pressurized to 25 Pa. In this situation there will be no pressure difference between
the ductwork and the inside space, therefore all the leakage measured will be between the
ductwork and outside.
Duct Insulation
For flexible duct, the insulation level is often printed on the outer jacket. Otherwise insulation
levels have to be estimated from the thickness and insulation type (for example glass fiber
blanket and batt is approximately R-3.7 per inch). We recommend R-8 for ducts outside the
conditioned space. R-8 ducts are rare and therefore most ducts will require additional insulation.
The most common method is to apply a duct wrap of glass fiber insulation. Other alternatives
include burying ducts in blown-in insulation – either restrained by cardboard forms or as part of
additional envelope insulation (e.g., insulation added to attic floors when ducts are lying on the
attic floor).
Spot Ventilation
Exhaust and supply fans in existing houses often have poor air flow due to bad installation (due
to highly restrictive ducting) and accumulated dirt. These fans should be replaced with quiet,
efficient fans that move the correct amount of air for their application – usually kitchen (100
cfm) and bathroom (50 cfm) venting. In some houses there may be more complex ventilation
systems, such as Heat Recovery Ventilators (HRV), that are also prone to the same problems.
Measuring the flows from these exhaust and supply fans can be accomplished using the same
methods as used for airflows at registers. Another issue is to ensure correct routing of exhaust
air. It is common to see exhaust fan ducts terminated inside attic spaces instead of going all the
way through the roof. This condition is important in heating climates because the moist indoor
air can condense on cold attic surfaces and lead to moisture problems. Kitchen range hoods tend
to be vastly oversized in order to compensate for restrictive ductwork. This causes noisy fan
operation, and can also cause a dangerous backdraft potential if there are natural draft
combustion appliances located in the conditioned space.
8
DRAFT – DO NOT QUOTE
Air Flows at Registers
Having the correct airflow at a register means that the HVAC system will be able to keep an
individual room or area of the house comfortable. Target airflows should be based on
engineering calculations using standard techniques such as those published by the Air
Conditioning Contractors of America (ACCA) in their Manual J and Manual D. These targets
should be combined with the results of discussions with the occupants. For example, if
occupants complain about the heating or cooling of an individual room, then the airflow to the
room needs to be investigated and changed. Measuring airflows at registers also allows us to
identify possible disconnected ducts. Residential register airflow techniques range from the use
of complex (but highly accurate) powered flow hoods, to simply timing how long it takes to fill a
plastic bag. For system balancing, detecting disconnects or setting register flows, most of these
techniques are acceptable (for a more thorough discussion of these measurement techniques see
Walker and Wray 2003).
Air Handler Flow
Air handler flows need to be checked because systems with too low a flow (<400 cfm/ton of
cooling in a dry climate, <350 cfm/ton of cooling in a humid climate or <12.5cfm/kBtu/h of
heating) tend to have poor efficiency for cooling or too high duct temperatures. In addition, for
cooling operation, low air flow leads to icing of coils, and periodic failure of the system until the
ice melts. Over time this often leads to premature failure of the compressor. For heating
operation, the low air flows lead to excessive conduction loses, furnaces operating on high limit
switches and dangerously hot furnaces and plenums. However for cooling systems in humid
climates, the airflow may be set deliberately low in order to increase the latent capacity of the
equipment. At the other end of the spectrum, systems with too high an airflow will tend to have
noise problems and poor moisture removal in humid climates. Air handler flows should be
measured using standard techniques such as pressure matching using an auxiliary fan and
flowmeter or flow plates inserted into the system. (e.g., ASHRAE 152, California Title 24 and
soon to be in ASTM E1554).
In this case study, we used the pressure matching technique, flow plates, and a research level test
procedure that measures air handler airflow using the constant tracer gas injection method. This
test involves injecting a known amount of tracer gas at the return grille, making sure that the
return run is long enough to mix the tracer gas evenly into the air so that there is a uniform
concentration of tracer gas within the duct, and (ideally) no tracer gas outside the zone. The
tracer concentration is measured in a supply grille, and the flow is determined by calculating the
volume flow of air necessary to create that concentration, with the known amount of gas
injected.
Filter Condition
Filter condition is determined by visual inspection. ASHRAE Standard 62.2 recommends a
minimum MERV rating of 6 for residential filters. These pleated filters are widely available in a
one inch width, which allows them to fit into standard filter slots.
9
DRAFT – DO NOT QUOTE
Ventilation Fan Power Consumption
Ventilation fan efficiency can be determined for many fans by visual inspection of the product
label, and cross-referenced with published performance values. However, many older fans will
not have published performance values and the only way to evaluate them is through diagnostic
testing.
Equipment Capacity
Nameplate capacity of equipment is determined by visual inspection of the label. When
performing retrofits, the correct equipment capacity should be determined using ACCA Manual J
to determine the heating and cooling load of the house and Manual D to size the equipment and
the ducts – including any improvements to the thermal envelope of the building such as
increased ceiling insulation or better windows.
Refrigerant Charge
Refrigerant charge has an optimum level for efficiency and capacity. Both over and under
charging reduces system performance, but undercharging is more common. In severe cases poor
system charge can lead to equipment failure. Refrigerant charge can be estimated using standard
industry test procedures: superheat or subcooling – depending on the manufacturers
recommendations for an individual piece of equipment. One difficulty with these methods is that
they cannot be performed in extremely dry weather or when it is cool. This restriction is
primarily an issue when installing systems in winter months as it means the systems cannot be
tested. When evaluating an existing system there are other indicators of charge problems – such
as frozen indoor evaporator coils – that can be used as diagnostic tools. A refrigerant sight-glass
can be used to visually monitor charge under various operating conditions. It is spliced into the
liquid line at the inlet to the evaporator coil. If the A/C is in steady-state operation and bubbles
appear, the system is under-charged. Charge is then added slowly until the bubbles disappear
and only liquid is visible.
The quantity of refrigerant charge in the system can be determined using the superheat method
(Consortium for Energy Efficiency, 2000). Superheat refers to the temperature difference
between the refrigerant leaving the evaporator coil and the saturation temperature of the
refrigerant at that pressure. The superheat test can be used on fixed orifice or capillary
expansion valve systems. It can only be used if the outdoor temperature and indoor temperatures
are within a certain range. Generally, indoor temperature should be above 70 degrees F and
outdoor temperature should be above 80 degrees F. If the superheat is too low it means that the
refrigerant is in liquid form throughout too much of the evaporator tubing, indicating that the
system is overcharged. If the superheat is too high it means that the refrigerant is changing into
a gas too early and the system is undercharged. In order to do the test the suction line refrigerant
pressure is measured and the saturation temperature calculated. Then the suction line refrigerant
temperature is measured and the two are subtracted to find the superheat. The actual superheat
must be compared to target superheat to determine if it is too high or too low. The target
superheat is calculated for the given indoor and outdoor conditions by using a superheat chart
provided by the air conditioner manufacturer.
10
DRAFT – DO NOT QUOTE
The subcooling method is similar to the superheat method, but it uses the temperature difference
between the refrigerant leaving the condenser and the condensing temperature of the refrigerant
at that pressure to determine if the system is adequately charged. This method is used only for
systems with thermostatic expansion valves (TXV). The superheat method doesn't work with
TXV systems because the TXV controls the refrigerant flow such that the superheat remains
constant (or within a range).
Condition of HVAC system
A simple evaluation of the age and visual appearance of the HVAC system can be a good
indicator for repair or retrofitting. An older system will almost certainly have less efficient
equipment than newer equipment and the degradation of air seals and flue systems with age
present good candidates for retrofitting. A system that shows dirty, rusting sheet metal, missing
duct tape, crushed fins, missing fasteners, dust marked open faced insulation, poor alignment of
flues or air ducts, missing flues or air ducts, missing insulation, etc., is showing signs of neglect
and is a good candidate for repair or retrofitting.
Window AC Units
Nameplate information is used to determine capacity and efficiency from manufacturer’s
information. The same observations of system degradation stated above are indicators for
replacement.
Multiple Systems/Zoning
For zoning, the following questions should be answered: 1) Are there multiple heating/cooling
systems in the house? Are they all the same age/condition – need to measure leakage/evaluate
insulation separately. Are the ducts connected correctly? 3) Is the system zoned? Check
thermostat locations and zonal separation. Check zone damper operation. Evaluate coil
bypass/multi-capacity operation for zoning.
Envelope Leakage
Most houses in the US have leaky building envelopes that result in excessive air infiltration,
uncomfortable drafts, and high energy bills. For many years weatherization programs have
worked at sealing building envelopes to cure these problems. Sherman and Dickerhoff (1994)
have compiled data from a large number of pressurization tests (over 10,000) in the United
States, and have found the average normalized leakage to be 1.72 for single family dwellings.
This study also showed an average reduction in Normalized Leakage (NL
2
) of 0.35 due to
weatherization. Buildings in the 70's and 80's were built tighter than earlier vintages, and some
had indoor air problems due to inadequate ventilation air. Ventilation air can be supplied
mechanically or naturally. ASHRAE Standard 62.2 sets the natural ventilation requirement at
2
NL is a method of accounting for the size of house in terms of its floor area and height when
comparing envelope leakage between houses
11
DRAFT – DO NOT QUOTE
0.35 ACH (on the order of NL=0.35), and the mechanical ventilation limit at 15 cfm per person
plus 50 cfm for each bathroom and 100 cfm for the kitchen. See also ASHRAE standards (62.2,
119 and 136) and other codes like California State Energy Code (Title 24).
Envelope leakage is measured using standard pressurization techniques (e.g., ASTM E779,
ASTM E1827, CGSB Standard 149.) The fan pressurization method involves closing all the
doors and windows, setting up a large fan in an opening (usually a door) in the building shell. A
pressure (usually between 20 Pa and 70 Pa) is maintained across the building shell, and the flow
through the fan is measured. ASTM E779-99 (1999) and E1827-95 (2000) give standardized
procedures for these tests. Sometimes a single point test is performed, generally at 50 Pa, or a
multiple point test is performed over a range of induced pressures. If a multiple point test is
performed, a curve of the form in equation 1:
()
n
QCP=∆ (1)
can be fit where Q is the flow through the fan and P is the pressure differential across the
building shell. The parameters C and n are characteristic of the particular building being
measured and are determined from the curve fit.
Often the leakage between various buildings is compared using CFM
50
, the leakage through the
building shell at 50 Pa. The information that would be more useful for energy use estimates is
the flow across the building envelope at normal operating conditions. This flow is constantly
changing since the pressure across the shell varies with weather conditions. To account for this,
a parameter called Effective Leakage Area (ELA) was developed, which is the area of an orifice
that provides the same flow as the building shell when subjected to a reference pressure
differential. The pressure differential that is commonly used in the US is 4 Pa because it is
typical of the pressure that building envelopes experience.
ASTM method E779-99 (1999) outlines how to determine building envelope leakage area
(including the calculation for ELA) using fan pressurization. Although fan pressurization does
not give a measurement of actual ventilation during normal building operation, it is a useful test
in order to compare the leakage area of two different buildings, to assist in identifying leakage
sources, and to determine the leakage reduction from an individual retrofit.
Ceiling, Wall and Floor Insulation
Envelope insulation needs to be estimated from observation of insulation type and thickness. In
retrofit cases walls can be insulated with blown in insulation if the existing stud cavity is empty.
Attics can be insulated with blown insulation that covers the joists to minimize thermal bridging,
and basements and crawlspaces can be insulated with fiberglass batts. There are some areas that
are unpractical to insulate in a retrofit situation such as under a slab on grade, or in wall cavities
that are already partially insulated.
12
DRAFT – DO NOT QUOTE
Windows and Window Shading
Because windows contribute significantly to the thermal load for houses this information is
required so that reasonable building load and equipment sizing calculations can be made. In
addition, a retrofit that includes added shading to south or west-facing windows in sunny cooling
dominated climates requires an estimate of window size and location. Window type is a good
indicator of the retrofit potential for the windows. Single-pane aluminum-frame windows have
poor thermal performance (and have high condensation potential) and are good candidates for
replacement whereas double pane vinyl-frame windows would not need to be replaced. Simple
observation of frames looking for signs of moisture damage will give valuable information
regarding not only window performance and likelihood of replacement – but also of high
humidity conditions in the house or the presence of water leaks in around the window openings.
Window retrofits are generally not justified in a retrofit program due to the large expense and
long payback time, however, window retrofits are often undertaken by home owners for other
reasons such as condensation resistance or aesthetics. Therefore there can be an opportunity to
combine energy efficiency gains from window changes together with changes in the HVAC
system.
Solar Radiation Control
The presence of radiant barriers, sunshades, or solar-reflective roofs means that these items (if in
good condition) that would normally be part of a retrofit do not need to be considered. Solar-
reflective roofing materials are commercially available for sloped-roof applications (white
asphalt shingles and colorful ceramic tiles) and for low-sloped or flat roofs (white acrylic,
elastomeric and cementitious coatings, and white, gray and blue thermoplastic membranes).
Refer to the LBNL Heat Island Group’s Cool Roofing Materials Database at
http://eetd.lbl.gov/CoolRoofs/
and the EPA Energy Star for Roof Products Qualifying Products
List at http://yosemite1.epa.gov/estar/consumers.nsf/content/roofbus.htm
. We do not
recommend painting or coating a conventional asphalt shingle roof white. The numerous seams
in the asphalt shingle roof make it possible for water to accumulate under the shingle edges,
particularly in humid climates. With a dark-colored shingle roof, water that has accumulated
evaporates the next time the shingle heats up. With a white-coated roof, the shingles tend not to
heat up enough to fuel water vaporization – leading to potential moisture damage.
Ceiling, Wall and Floor Construction
The type of wood or steel framing or the use of other materials restricts the places that HVAC
ducting or plumbing can be run.
Moisture
Visually inspect for signs of moisture such as wetness, mold, mildew, variations in color and
texture, and dimensional/structural problems. Note that visual observation is only good for
existing problems in visible surfaces and cannot evaluate the potential for future problems,
problems buried within the construction of the home. Also, be aware of any special odors
associated with moisture. Electrical inspection by impedance scanning and conductance probing
13
DRAFT – DO NOT QUOTE
can be used to detect subsurface problems before they are visually apparent. One of the
objectives of the retrofit packages is to have HVAC systems that are effective at controlling
moisture and do not increase the potential moisture problems for the building envelope.
Occupant Survey
Ask occupants to report problems (comfort. high bills, condensation, mold, etc.) and ask about
important lifestyle activities that can significantly change building loads and the times that the
house needs to be conditioned. The following are some typical questions that might be asked:
How many people live in the house? More occupants indicate that the chances for
humidity and other Indoor Environmental Quality (IEQ) problems will be greater
(therefore a tight house envelope will require the addition of mechanical ventilation).
Are there any pets? Like human occupants pets are a source of moisture and odors. This
means that adequate ventilation is required – with more ventilation required for more or
larger pets. Fish-tanks are a source of humidity – particularly if they are large and/or
uncovered. Exotic pets may have particular temperature and humidity requirements that
make for unusual building loads – check with the homeowner. Pets may also restrict the
use of setback or setup programmable thermostats.
High Energy Bills? High energy bills can be a good indicator of HVAC system
problems, and the potential to perform envelope upgrades makes more financial sense if
there is the potential to save a lot of money.
Diagnostics and Screening Results from Four Houses
Four houses in a mixed-dry or hot-dry climate within 50 miles of the San Francisco Bay were
surveyed using the diagnostic screening approach (Table 1) and one was selected for retrofit.
Our objective was to select the house with the greatest potential for improvement from the
retrofits. The four houses will be referred to by location: Concord, Moraga, Castro Valley and
Larkspur. The key screening results for the four residences are shown in Table 2. More details
of the screening results can be found in the tables of Appendix A.
14
DRAFT – DO NOT QUOTE
Table 2. Comparison of Initial Screening Results of the Four Residences.
Diagnostic Units Concord Moraga Castro Valley Larkspur
duct leakage at
operating - supply
L/s (cfm) 15 (48) 111 (365) 33 (109) 37 (122)
duct leakage at
operating – return
L/s (cfm) 100 (328) 92 (302) 16 (54) 63 (207)
duct leakage at
operating –
supply
6% 22% 9% 10%
duct leakage at
operating – return
41% 18% 5% 17%
air flows at
registers (sum
supply)
L/s (cfm) 238 (779) 380 (1246) 251 (823) 342 (1120)
air flows at
registers (sum
return)
L/s (cfm) 133 (435) 433 (1419) 290 (980) n/a
air handler fan
flow (cooling
mode)
L/s (cfm) 245 (804) 502 (1646) 353 (1159) 371 (1216)
superheat test
(difference from
target superheat)
°C (°F)
8.9 (16) -7 (-12.7) 7 (12.6) 0 (0)
refrigerant charge undercharge
d
overcharged undercharged correct charge
measured
equipment
capacity
W (btu/h) 8089 (27600) n/a n/a 9847 (33600)
heat load Man. J W (btu/h) 28,741
(98,070)
21,951
(74,900)
12,690
(43,300)
n/a
cool load Man. J W (btu/h) 12,305
(41,985)
17,203
(58,700)
8,675 (29,600) n/a
rated capacity of
existing cooling
equipment
W (btu/h) 11,606
(39,600)
22,859
(78,000)
12,749
(43,500)
10,551
(36,000)
system oversize
ratio
0.94 1.33 1.47 n/a
envelope leakage m
2
(in
2
) 0.179 (278) 0.229 (355) 0.174 (269) 0.219 (340)
ceiling insulation R 26 17 25 inaccessible
The Castro Valley house was eliminated because the building shell and ducts were the tightest of
the four houses. It was a difficult choice between the remaining three houses. They all had
15
DRAFT – DO NOT QUOTE
leaky building shells, and leaky duct systems. The Moraga house was complicated by the fact
that there were two HVAC systems. One of the HVAC systems was 15 years old and leaky, and
the second was newer (only 7 years old) and had very little leakage. Both systems were slightly
overcharged. The Larkspur house had a leaky shell and leaky ducts, but it had a condensing
furnace with 3 ton air conditioner, which performed almost at the rated 3 tons (2.8 tons on the
day that we tested it). The shell leakage would be impossible to fix at this house since the
ceilings were all tongue and groove boards, which are difficult to seal without changing the
desired appearance.
The Concord house was selected for the case study because it showed the greatest potential for
improvements. The house was a 27-year old single-family two-story dwelling of approximately
2500 ft
2
with attached garage, and was cooled and heated by it’s original central gas furnace/air-
conditioning system located in the garage, with ducts primarily located in the attic and garage.
The roof was constructed with colored ceramic tiles on a sloped plywood deck, over a naturally
ventilated and unconditioned attic, with fiberglass insulation (R-26) between the 2 by 8 joists on
16” centers. A floor plan of the test residence with existing ducts is shown in Figure 1 and
exterior views are displayed in Figure 2.
The house had the following combination of problems: low-efficiency heating and cooling
equipment (Figures 3 and 4), leaky and poorly insulated ducts (Figure 5), and a leaky exterior
envelope (also in Figure 5). In addition, the air handler, furnace, cooling coils and most of the
duct system were located outside the conditioned space in the garage and attic. A few major
components of the shell leakage were easily identified in this house: several large mechanical
chases that were open to the attic, as well as a building cavity return that was open to the garage
and the attic. The HVAC system was undercharged and operating at only 2/3 of its rated
capacity. We suspected that the refrigerant line had a leak due to the low charge in the system.
Also, the homeowner reported problems in cooling the upstairs of the house.
16
DRAFT – DO NOT QUOTE
Legend
return
supply
C
l
o
s
e
t
Master Bedroom
Closet
Bath
C
l
o
s
e
t
Bedroom
Bedroom
Closet
Closet
Office
Bath
Laundry
Garage
Bath
Family Room
Playroom
Breakfast
Kitchen
Dining
Living
Figure 1. Floor plan of case study residence with existing ducts. Upper level on top.
17
DRAFT – DO NOT QUOTE
Figure 2(a). Exterior front view of
Concord test residence.
Figure 2(b). Exterior side view of
Concord test residence showing gable
attic vents and existing outside AC
condenser unit placement [lower left].
Figure 2(c). Exterior rear view of Concord test residence.
18
DRAFT – DO NOT QUOTE
Figure 3. Existing HVAC equipment in garage [left to right: return plenum, air
handler, furnace, cooling coil and supply ducts].
Figure 4. Existing AC condensing unit outside near rear of garage showing poor
condition of unit [heavily corroded sheet metal fasteners and significant dirt/dust
accumulation blocking flow into coil] and subsided slab [stressed refrigerant tubing and
subsequent leakage of refrigerant].
19
DRAFT – DO NOT QUOTE
Figure 5(a). Existing attic floor
showing insulated metal ducts and
insulation coverage.
Figure 5(b). Existing insulated metal
duct splitter and insulation coverage.
Figure 5(c). Existing supply duct run from garage to attic showing large openings in
attic floor plane [above stairway dropped ceiling].
Hole into interior partition
Figure 5(d). Existing return cavity between
foundation and lower level floor.
Crack between concrete
foundation and sheet
metal
20
DRAFT – DO NOT QUOTE
Key Screening Results: Case Study Residence
The key screening results and occupant issues for the Concord test residence are listed below and
the pre- and post-retrofit diagnostics are shown in Table 3.
1. Leaky envelope (10 ACH
50
equivalent to an NL of 0.9).
2. Dirty and leaky ducts – 48 cfm supply and 328 cfm return (47% of air handler flow supply
and return combined.
3. Low air handler flow only 800 cfm (rather than the 1400 cfm expected for a 400 cfm/ton, 3.5
ton nominal system
3
).
4. Refrigerant charge 25% low.
5. Dirty condenser cooling coil with crushed fins.
6. Building in hot-dry moderate climate.
7. Building, furnace/air handler/ac are >25years old.
8. System is easily accessible in garage.
9. Wall cavity ducts are inaccessible between floors.
10. Poor air distribution with not enough cooling upstairs.
11. Ineffective and/or non-operative electrostatic air filter.
12. Condenser unit is corroded.
13. Condenser unit is poorly sited in sun and on subsided concrete pad.
14. New (installed in 1998) ceramic tile roof in excellent condition.
15. Insulated walls.
16. Single pane windows with aluminum frame.
Occupant Issues:
1. Reported distribution of cooling is not adequate: upstairs is hotter than downstairs when
cooling.
2. Occupants have sensitivity to airborne particles – so the HVAC system requires good
filtration.
3
With the low air conditioning capacity (mostly due to undercharge) the low air handler flow is
less of a problem, because the target air handler flow would be only 900 cfm (for the 2.3 ton
measured capacity). This condition shows how coincident problems can mask each other. These
two problems result in poor HVAC efficiency, but will not show up as system problems such as
frozen coils or failed compressor.
21
DRAFT – DO NOT QUOTE
Table 3. Diagnostics Summary Sheet of Pre- and Post-Retrofit Comparison at the Test
Residence.
Diagnostic Units Pre Post
duct leakage at operating – supply cfm 48 n/a
duct leakage at operating – return cfm 328 n/a
duct leakage at operating – supply 6% n/a
duct leakage at operating – return 41% n/a
duct leakage at 25 Pa – supply cfm 86 60
duct leakage at 25 Pa – return cfm 786 136
duct leakage at 25 Pa – supply 11% 4%
duct leakage at 25 Pa – return 98% 9%
air flows at registers (sum supply) cfm 779 1458
air flows at registers (sum return) cfm 435 n/a
air handler fan flow (cooling
mode)
cfm 804 1525
refrigerant charge undercharged correct charge
measured equipment capacity btuh 27600 n/a
heat load Man. J btuh 98070 82771
4
cool load Man. J btuh 41985 38150
rated capacity of existing cooling
equipment
btuh 39600 36000
system oversize ratio 0.94 0.94
envelope leakage in
2
278 208
ceiling insulation R 26 54
ACCA Manual J calculations were performed on a room-by-room basis to estimate heating and
cooling loads. Based on the Manual J calculations the airflow required by each room was
determined as shown in Table 4. The existing air flow was compared to the ideal airflow
calculated by Manual J to see if there were existing problems with the distribution throughout
the house. The downstairs of the house had slightly lower airflow than required, and the upstairs
had slightly higher, except for the master bedroom, where the airflow was too low. The
homeowner had reported that the cooling was not adequate in the upstairs of the house, which
indicates that there is a large conductive loss in the supply ducts, or that there is some other
reason that the upstairs is getting too hot (for example, infiltration of hot attic air into the upper
part of the house.)
4
Retrofits included adding ceiling insulation and air sealing the building shell, which reduced
the heating and cooling loads.
22
DRAFT – DO NOT QUOTE
Table 4. Measured and ACCA Manual J Calculated Register Airflow for the Test Residence
in the Pre-Retrofit Condition.
Location of register
Measured flow
[cfm]
Calculated flow
(using ACCA Manual J)
[cfm]
dining/living (near kitchen
door) 101 156
dining/living (under piano) 55 74
game room (bathroom) 126 131
game room (laundry) 140 131
bathroom 28 8
master bedroom 36 102
master bedroom (near closet) 32 51
master bathroom 28 15
bedroom (front of stairs) 67 40
bedroom (corner) 60 45
office 76 43
girls bathroom 29 9
Manual D calculations to size the new ducts to deliver the required airflow out of the registers
are shown in Table 5. The retrofit included a single speed air conditioner with a zoned duct
system. In all cases except one the size was adequate for low velocity (when both upstairs and
downstairs were calling for conditioning), but there were some cases where the size was too
small for the high velocity case (when just one zone was calling). There were a total of 18 ducts
sized, 8 were too small on high velocity and one was also too small on low velocity. Of these 8
that were too small, 5 were metal ducts retained from the original duct system. It was not
possible to change these 5 metal ducts without considerable expense. Thus 3 of the ducts that
the contractor installed were too small for the high velocity case.
23
DRAFT – DO NOT QUOTE
Table 5. ACCA Manual D Calculation of Duct Airflows for the Test Residence.
Design flow
(cfm)
Post retrofit
duct
Velocity at
design flow
(fpm)
run trunk length
(ft)
low high type size
(in.)
low high optimum
size (in.)
notes
1 ST1 2 237 386 metal 6 1100 1800 10 a
2 ST1 3 111 181 metal 6 550 900 8 a
3 ST2 6 187 305 metal 8 550 850 10 a
4 ST2 16 187 305 metal 8 550 850 10 a
5 ST2 1 14 23 metal 4 180 250 4
6 ST5 7 158 403 flex 9 350 900 12 b
7 ST5 13 79 202 flex 8 300 600 8
8 ST5 10 25 63 flex 5 200 500 5
9 ST4 3 59 150 flex 7 <200 500 7
10 ST4 12 69 175 flex 7 <200 600 7
11 ST3 4 66 167 flex 7 <200 600 7
12 ST3 4 16 40 flex 4 <200 450 4
ST1 18 348 567 flex &
metal
14 <200 500 14
ST2 25 389 634 flex &
metal
12 400 750 14 c
ST3 5 82 207 flex 8 <200 600 8
ST4 19 128 325 flex 8 400 900 12 d
ST5 14 104 265 flex 12 <200 400 12
ST6 2 201 328 metal 8 400 900 12 a
Table 5 notes:
Friction rate design value: 0.067 IWC/100 ft.
a: can’t change, b: can change, but homeowner wants smaller flow, c: difficult to change
d: can change
Retrofit Selection
System Sizing
A primary consideration for retrofits should be to downsize replacement systems. Sizing is a
common problem, and generally systems are oversized which leads to short run times. Short
runtimes are less efficient, have poor humidity control, have poor mixing throughout house
(more stratification), provide less comfort, and can result in premature compressor/fan failure.
If the duct system is improved with increased insulation and reduced leakage, and/or the building
load is reduced through envelope changes, then the equipment capacity can often be downsized.
Additionally, most residential systems are already oversized and would benefit from correct
sizing to reduce cyclic losses and improve part load humidity control. In many climates
24
DRAFT – DO NOT QUOTE
requiring significant cooling sizing for cooling leads to extreme heating oversizing. This
oversizing is because the air handler flow is determined for cooling operation. There is a limited
range of furnace capacities associated with each air handler, and these heating capacities tend to
be much higher than the usual factor of 1.7 times the heating load. If a house can be cooled with
a lower capacity system, it therefore has the synergistic benefit of better matching the heating
load. The major benefits of correct heating sizing are less cyclic losses (when hot ducts are left
to cool at the end of the heating cycle) and less switching on and off that can irritate occupants.
The house air will also be better mixed because a smaller system will have longer run times.
The extra cost of non-HVAC equipment related retrofits can often be compensated for by the
reduced cost of a lower capacity heating and cooling system. This strategy is being used
successfully in new construction (for example by various Building America Programs
http://www.eere.energy.gov/buildings/building_america/system.shtml
). Downsizing should also
include the reduction of duct diameter and register replacement in order maintain throw from
registers for good room mixing. Successful examples of these tradeoffs for new construction are
illustrated in Building Science case studies:
http://www.buildingscience.com/buildingamerica/casestudies/default.htm
In mixed-dry, hot-dry climates where air moisture is low as in California, the evaporator cooling
coil can be over-sized to increase the sensible heating ratio. The contractor in this case study
matched a 3 ton outdoor coil with a 4 ton indoor coil. The airflow over the indoor coil is set
based on the capacity of the outdoor coil. We were able to downsize the cooling system by 14%
(0.5 tons) by reducing duct losses and improving the building envelope. The heating input
capacity was also reduced from 120 kBtu/h (35 kW) to 70 kBtu/h (20.5 kW). Part of the reason
for these reductions is the significant improvement in efficiency for the newer systems compared
to the older ones.
Keeping the Design Simple
A key aspect of the keeping the design simple idea is to carefully examine the existing HVAC
system to see what can be saved or salvaged for reuse. Parts of the system that are functioning
well should be retained as much as possible. A good example is sheet metal ducts that are
normally replaced with new flexible duct by contractors. The sheet metal ducts have much less
flow resistance (and therefore less noise) and superior durability. If they can be sealed (if leaky)
and insulated (if un-insulated), they should be retained if possible. In this case study we had
planned to retain the sheet metal ducts in the attic and we had extensive discussions with the
contractor about this issue. Unfortunately, the ducts were replaced on a day when we were not
present at the job-site and the contractor removed the sheet metal ducts and replaced them with
new flexible duct. The reason given by the contactor was that they were concerned that the
existing ducts were undersized. This is an important misconception because flex ducts have very
high flow resistance and restrict the flow more than equivalent sheet metal duct (Abushakra et al.
2002.) The contractor in this case did increase the size of the ducts when he replaced the metal
duct with flex duct.
25
DRAFT – DO NOT QUOTE
Durability
Retrofits need to be durable so that the benefits will still be in place 25 to 50 years from now.
Examples of durability concerns include not using cloth duct tape (due to thermal degradation,
see Walker and Sherman 2003), and avoiding moisture problems. This retrofit case study
addressed the durability issue by having better equipment sizing for better humidity control,
more ventilation (by adding an economizer) for better indoor IAQ and humidity control, and use
of good duct sealants to prevent entry of humid outdoor air.
Retrofit Package
Based on the results of the screening, we chose a retrofit package centered on four areas of the
building system for a mixed-dry, hot-dry climate: [1] Building Envelope – tightened [2] Attic
Floor Insulation -additional blown-in fiberglass [3] HVAC Equipment - new high efficiency
downsized system with an economizer and zone control [4] Ducts - sealed and covered with
additional insulation. The following 15 retrofit actions have been selected because they are all
off the shelf existing technologies, and cover a range of energy, comfort and durability issues.
Building Envelope Retrofit
1. The sealing of the envelope was very successful – mostly because this particular house had
significant large leaks that we were able to access. We sealed over 1200 cfm of leakage at 50 Pa
(about one-quarter of the total leakage). The sealing included: air-sealing of the attic floor plane
(2 large chases) as shown in Figure 6, the installed Therma-Dome (attic pull-down staircover),
leaks between the old return (which communicated with the garage) and the conditioned space,
and plumbing penetrations in the conditioned space.
Sealed Chase
Open chase
before sealing
Figure 6. Sealing cavities connecting the house to the attic. Foam is a useful sealing tool for
small holes and cracks at building component intersections (left). Duct board insulation can
be used to block off large open areas (center and right)
26
DRAFT – DO NOT QUOTE
2. Unused cavity returns were sealed (after HVAC installation).
HVAC Equipment Retrofit
The selection of which equipment to use was fairly straightforward because the contractor
already installs high-efficiency systems with economizers in new construction. Therefore the
contractor was able to give us several options from different manufacturers that used condensing
furnaces (95% AFUE), high-efficiency air conditioners (SEER 14) and featured air handlers that
remain efficient at lower speeds.
The existing low-efficiency standard central forced-air air-conditioner and furnace components
were replaced with a high-efficiency system by a local contractor who was willing to collaborate
with us by installing a “non-standard” system. The contractor also installed a programmable
temperature-controlled economizer and a two-zone thermostat controller. The furnace, air-
handler, cooling coil and plenums were relocated from the garage to the attic as shown in Figure
7. A pull-down staircase was added for better attic maintenance access (this access is important
for filter changes. In particular, this household had occupants who were sensitive to IEQ issues
and needed good air filtration). We would have preferred not to use the attic, because its
temperature extremes lead to poor duct and equipment efficiency. In this case, there was
insufficient room in the garage for the economizer ducting and the attic was the only place to put
the air handler if we wanted an economizer.
Figure 7(a). New cooling and heating
equipment on attic floor [right to left:
return plenum, air filter, air handler,
furnace, cooling coil and supply plenum].
Figure 7(b). New cooling coil with
condensate drains and top of furnace with
combustion air intake and exhaust.
27
DRAFT – DO NOT QUOTE
The heating and cooling equipment capacity were sized using the ACCA Manual J
calculation and engineering considerations derived from the monitored data. The
selected heating capacity of the two-stage furnace was 66.9 kBtu/h (19.6 kW) and 46.4
kBtu/h (13.6 kW)and the cooling capacity of the AC was 3 tons (36 kBtu/h (10.5 kW)).
As shown in Table 3, the post-retrofit Manual J calculation of the heating load was 82.8
kBtu/h (24.3 kW) and the cooling load was 38.2 kBtu/h (3.2 tons) (11.3 kW), a reduction
of 15.3 kBtu/h (4.5 kW) in heating load and 3.8 kBtu/h (0.3 tons) (1.1 kW) in cooling
load. The system is slightly undersized according to ACCA Manual J, however, we
know that the ACCA calculations lead to slightly oversized systems. It is interesting that
the contractor recommended a system in this case that is undersized rather than oversized
since we generally think that contractors oversize systems in order to reduce their
callbacks. As shown in Appendix C, the pre-retrofit operating cooling capacity was 30.6
kBtu/h and the sensible portion was 21.6 kBtu/h (6.5 kW), which results in a sensible
heating ratio of 0.71.
3. A downsized Amana Prestige Ultra RCE high efficiency split system air-conditioning
package consisting of a remote condensing unit (RCE36A2D) and an over-sized cooling
coil (Aspen BBL48+X2: 48 kBtu/h (14 kW)) was selected with the following
specifications: 36,000 Btu/h (10.5 kW) nominal capacity, 0.73 SHR and 14 EER at
standard conditions and 1200 cfm. A Seasonal Energy Efficiency Ratio (SEER) of 12 or
greater is the standard for an Energy Star qualifying central air conditioner. The
cooling coil was relocated from the garage to the attic. The condensing unit was
relocated from a sun-exposed area behind the garage with unstable soil to a shaded area
on the opposite end (north) of the house with a new slab on stable soil as shown in
Figure 8. The old unit was placed on a very poor slab that has tilted considerably.
Figure 8. New AC condensing unit on shaded side of house.
4. The existing gas furnace and air-handler were removed from the garage and replaced
with a downsized high-efficiency unit in the attic to allow for the use of an economizer.
An Amana Air Command 95IIQ GUVA horizontal flow variable speed two-stage gas
furnace (GUVA070BX40) with a ¾ hp blower motor was selected with the following
specifications: 66,900 Btu/h (20 kW) high-fire rate output, 46,400 (13.5 kW) Btu/h low-
fire rate output and 95.5 AFUE at standard conditions. In direct vent mode, this furnace
28
DRAFT – DO NOT QUOTE
is designed to draw combustion air from the outside; however it was installed to draw
combustion air from the attic. The furnace flue pipe was installed through the roof as
shown in Figure 9. The air handler is equipped with a General Electric ECM
(Electronically Commutated Motor) (2.3 Series). These motors are advertised to be 20%
more efficient than a standard motor at full load and 30% more efficient at low speed.
The DC design allows the fan speed to be easily adjusted over a large continuous range
of speeds. The new system used a control strategy that slowly increased the air handler
speed at the beginning of each cycle. Another advantage of these motors is that the torque
does not change much with an increase in load so the fan is able to overcome ductwork
resistance, and deliver a given airflow regardless of duct restriction.
5. Installed improved air filtration with a 4” (100 mm) pleated MERV-11 air-filter at air-
handler inlet. An AirGuard PowerGuard pleated panel filter was selected and installed
(20x25x4 nominal size and 23.5 ft
2
(150 cm
2
) of gross media area). The filter qualifies
for MERV-11 Value per ASHRAE Standard 52.2. Inspection by LBNL of the filter
installation revealed a 1” (25 mm) air gap at the top between the filter and slot, which
allowed air to bypass the filter. The contractor was contacted to install the filter properly
and eliminate the space to force all the air through the filter.
6. Installed “Smart Vent” (fresh air ventilation controller model SV10) temperature-
controlled economizer in attic (through the roof – the economizer roof-vent is shown in
Figure 10). The economizer monitors indoor and outdoor temperatures. When the set
temperature difference is met the fan is turned on and a vent damper is activated allowing
filtered outside air to cool the house. The damper for the fresh air inlet was located in the
upper hallway return. The damper was designed so that when it opens the outdoor air
inlet, it automatically closes the return air pathway through the upper hallway return
grille. Another damper was installed in the return duct from the downstairs part of the
house to close off this return air pathway when the fresh air inlet opens. A pressure relief
damper also opens (to the attic) during economizer operation to prevent pressurization of
the house. The economizer has 3 settings: On, Off, and Auto. On means the outside vent
opens and the air handler fan turns on (in "fan on" mode), independent of what the inside
and outside conditions are. Off means the vent closes and the air-handler fan is
controlled by the zone thermostats. Auto means that the vent will open if the outdoor
temp is 4ºC (8°F) lower than indoor temperature, and if the smart vent thermostat
setpoint is at least 0.5ºC (1°F) lower than the indoor temperature (measured at the return).
To adjust the indoor/outdoor differential condition (factory set at 4°C (8°F)) the rheostats
in the control box in the attic can be adjusted. When the outside air is cooler than the
indoor air (usually at night or in shoulder season), the economizer will use the air outside
to cool the house. The conditions for the economizer to operate are that the house
temperature has to be 2ºC (4ºF) greater than the outside temperature and that the
economizer must turn off if the indoor temperature drops to 17ºC (62.6ºF).
29
DRAFT – DO NOT QUOTE
Figure 9. Installed roof mounted
furnace flue pipe.
Figure 10. Installed roof mounted
economizer vent.
30
DRAFT – DO NOT QUOTE
7. Installed ZTE two-zone control system for separate upstairs and downstairs control and
improved occupant comfort.
Duct Retrofit
The retrofit duct schematic for the supply side of the systems is shown in Figure 11.
8. Closed existing return because it was leaky and no possible way to seal it. Installed a larger
upstairs return in new location (upstairs hallway ceiling) to assist in reducing temperature
stratification. Installed a smaller downstairs return in new location (in the wall at the bottom of
the stairway).
9. Replaced metal ducts in the attic with new R-4 flexible ducts because the contractor thought
they were undersized. Initially the contractor hung the flex-ducts from the attic ceiling with
smooth bends as shown in Figure 12. However, we planned to add insulation over the ducts and
wanted them installed directly on the attic floor plane. The contractor then placed the new ducts
on the floor per our request in Figure 13, but unfortunately did not take the time to lay the ducts
with smooth bends as shown in Figure 14.
10. Added R-8 fiberglass wrap to R-4 insulated attic flex ducts for a total insulating value of
about R-12 as shown in the right hand photo in Figure 15.
11. Added at least 4” (R-12) of blown-in fiberglass insulation to R-4 attic flex ducts for a total
insulating value of R-16 as in Figure 15.
12. Added 4” of fiberglass wrap to R-4 insulated garage ducts for a total insulating value of
about R-16.
13. New plenums were interior insulated with 1” (R-4) of fiberglass ductboard.
14. Our goal was to seal supply and return ducts to less than 10% of air handler flow, which is
the standard for best practices (e.g. Energy Star Ducts). The ducts were sealed using manual
methods and with the aerosol method developed by LBNL. But we did not meet this
specification (14% return leakage, 9% supply leakage) due to excessive leakage at the
economizer outdoor air damper. Figure 16 shows the duct sealing using foam of the downstairs
return. The aerosol method internally seals leaky ducts by injecting an aerosol sealant into the
duct system as shown in Figure 17.
31
DRAFT – DO NOT QUOTE
Air Handler
Closet
Bath
Closet
Bedroom
Bedroom
Closet
Closet
Office
Bath
Master Bedroom
Closet
Figure 11. Retrofit duct schematic of case study residence. Changes were only made to the
upper level ducts. The return was modified as well, and is not shown.
32
DRAFT – DO NOT QUOTE
Figure 12. Initial installation of ductwork hanging from ceiling of attic [right shows smooth
bend out of plenum].
Figure 13. Second and final installed retrofit of ductwork after being moved to attic floor and
before adding insulation.
Figure 14. Second and final installed retrofit of ductwork after being moved to attic floor and
before adding insulation showing tight bends.
33
DRAFT – DO NOT QUOTE
Figure 15. Attic floor with flexible ducts and additional blown-in insulation over ducts. The
yellow material is fiberglass wrap insulation.
Figure 16. Duct sealing of downstairs return using foam. [the sensor that is visible is a relative
humidity sensor, there is also a temperature sensor in the duct].
Figure 17. Aerosol apparatus for duct sealing in garage.
34
DRAFT – DO NOT QUOTE
Attic Retrofit
15. Added 5” (R-17 R3.4/inch) of blown-in fiberglass insulation to R-37 attic floor plane for a
total insulating value of R-54 as shown in Figure 15.
Issues with Zone Control
A 2-zone HVAC control system with independently controlled motorized dampers was installed.
Each zone (second and first floors) has a thermostat with independent cooling and heating
control. The control unit also has an automatic re-circulating fan cycle that can be activated to
alternate between heating and equalizing cycles. Simplified and detailed wiring diagrams for the
HVAC system are displayed in Appendix E
The two thermostats have On, Off, and Auto modes. There is also the thermal equalizer mode
which triggers only when the upstairs is hotter than the downstairs - the upstairs supply damper
closes and the fan turns on to circulate the hot upstairs air to the downstairs when the downstairs
is calling for heat. With this system a heating call takes precedence over a cooling call, and the
control will satisfy the heating demand first before cooling. The zone control does not control
the fan speed. This is a problem since the fan is set to provide proper airflow when all of the
registers are receiving air. When the zone control activates only one zone then the full volume
of air is forced through approximately half the number of registers. This causes uncomfortable
airflow from the registers and excessive noise. We talked to the control manufacturer to see if
they were addressing this issue, and they told us that their next model of zone control would
integrate the zone control and the economizer, and would be able to modulate the fan speed
when used in conjunction with an ECM fan. They are currently working with one of the large
air handler manufacturers to work out how the controller would integrate with the air handler.
The equalizer function works off a timing sequence for furnace run time and equalizer run time.
The control can be set to run the furnace for 8,12,16, or 20 minutes and the equalizer to be "off"
or run for 6, 8, or 10 minutes. One note on the equalizer timing is that some furnaces will go
through a "cool down" during equalizer run time that will extend the above times by the furnace
"cool down" time. The control will not engage the equalizer function unless the furnace run time
is greater than 5 minutes.
Here is an example zone control sequence: Both thermostats call for heat. Both dampers open
and heat is given to both zones. If one thermostat indicates that its zone is warm enough that
zone damper will close and the control will run the furnace until either the other zone satisfies or
the furnace run time expires. If the furnace run time is greater than 5 minutes and total furnace
run time has exceeded the furnace run time setting of the control, the control will engage the
equalizer mode. The equalizer mode disengages the heating call, shutting down the burners of
the furnace, closes the upstairs damper and enables the furnace fan. Depending on furnace
manufacture the fan will run in either high or medium speed. This feature doesn't operate in
cooling mode.
In the case study system, there are two zones electrically controlled by dampers in the supply
plenum. We thought that the fan speed would be reduced when only one zone called for cooling
35
DRAFT – DO NOT QUOTE
so that the fan wouldn't be trying to force the full airflow through half of the ducts. However, in
the actual installation, the fan speed was the same no matter if one or two zones were calling for
heating or cooling. We found that the control system was operating as designed, and there is no
provision in the control system to change the fan speed when one zone shuts down. We have
consulted with the zone controls manufacturer (who is neither the contractor nor the equipment
manufacturer) to see if they had plans to improve their design, and they told us that they have a
new controller ready for beta testing that will digitally interface with an ECM fan motor. This
interface will allow the zone controller to determine the air-handler fan speed so that the speed
can be lowered when just one zone is calling for heating or cooling.
In doing this modification to the zone controller there is the potential to cause problems in either
heating or cooling mode. In heating mode, as airflow is reduced the temperature rise across the
heat exchanger increases. If the temperature rise is too high then the heat exchanger will cycle
on the high limit switch, if it is too low then condensation can form inside the heat exchanger.
Both of these results can cause damage to the heat exchanger. The range of temperature rise for
our installed unit is 30-60°F (15 to 30°C). In cooling mode, if there is too little airflow across
the evaporator coil, the moisture that condenses on the coil may freeze, causing an insulating
layer between the coil and the air stream, or in a severe case, liquid refrigerant may make it's
way back to the compressor, causing damage to the compressor. The problem of too much
airflow in cooling mode will not damage the equipment, but can cause excessive noise, which is
bothersome to the occupants (which is how the unit currently operates).
A potential solution is to put a temperature sensor in the supply plenum, and one inside the
return grille to measure the temperature rise in heating mode, and the supply air-stream
temperature in cooling mode. If these temperatures fall outside of a given range, then the
controller will adjust the airflow to compensate. This control system will probably work better
with multi-stage equipment than it will with single stage because the capacity of the equipment
will be able to be adjusted to match either a small load (a single zone calling for conditioning) or
a large load (all the zones calling for conditioning.) So in our test house, this lowering of airflow
is probably easier to do in heating mode, due to the heating staging available with the new
furnace.
At the house in this study, the cooling speed tap is set on "C", nominally 1100 cfm, and the
heating is set on tap "B", nominally 1325 cfm in high stage and 920 cfm in low stage (69% of
high stage heating). The flow in fan-only operation is nominally 616 cfm (56% of high stage
cooling). The zone controller, which takes input signals from the thermostats and sends output
signals to the air handler, uses a combination of furnace run time and temperature difference at
the thermostat (difference between measured temperature and setpoint temperature) to control
the number of heating stages that are brought into operation. If the delta T at the thermostat is
3°F or more, after 5 to 6 minutes of single stage operation, then the second stage of heating is
called on. We are collaborating with the controls manufacturer to implement a more
complicated strategy where the second stage of heating is not allowed to come on if only one
zone is calling for heating.
36
DRAFT – DO NOT QUOTE
Retrofit Costs
The total cost of the HVAC and building envelope retrofits were $13,675. The cost breakdown
from the HVAC contractor was $11,201 for materials and labor as shown in Appendix F to
upgrade the HVAC, add an economizer, add the zoned system, add a duct return from first floor
and add a pull-down staircase; $600 to aero-seal the ducts; $990 to add R-11 fiberglass to the
attic floor and insulate over ducts with fiberglass; $325 to insulate the garage ducts; $300 in
labor and $150 in materials to air-seal attic floor plane; $109 for an attic stair cover.
Retrofit Effectiveness
The Concord test house was monitored during the pre-retrofit period for about eight weeks from
July 12
th
through September 18
th
2002. The house envelope and HVAC system were then
retrofitted. Data collection then resumed on October 2
nd
for the post-retrofit period. The
weather at this point was too cold for the cooling system to operate, so we were unable to get
post retrofit cooling system data.
The instrumentation and measurements are described in detail in Appendix B.
Examination of the 10-second data shows the total AC power increased with increasing outdoor
air temperature at 69W/°C and likewise with compressor motor power at 70W/°C. However, the
condenser and air-handler fan power decreased with increasing outdoor air temperature at –1.2 &
-0.3W/°C. The Energy Efficiency Ratio (EER) decreased at -0.1/°C of outdoor air temperature
and cooling capacity decreased at -0.3kBtu/h/°C.
Duct leakage was measured at 13% (196 cfm
25
), where the Energy Star leakage requirement
of < 10% (150 cfm
25
) total leakage is the target. The leakage breakdown is 4% (60 cfm
25
)
supply - almost all in the cabinet and supply plenum and 9% (136 cfm
25
) return - almost all
of which is in the economizer damper box. The economizer housing was damaged during
installation, where nearly 136 cfm25 of leakage was measured.
1. The air filter has a 1” bypass between the top of the filter and the sheet metal housing.
This problem was mentioned to the contractor, and it was fixed with a piece of sheet
metal. Unfortunately, the sheet metal fix was installed on the opposite side of the filter
from the sheet metal fix at the back of the filter, so the bypass still exists.
2. It seems the contractor does not weigh charge, the condensing unit comes pre-charged
and no more was added. The contractor normally would check the charge with a
superheat test, but the weather was not warm enough to do one in this case.
3. The tension in the springs of the zone selection dampers was not adjusted so that they
open all the way against pressure.
37
DRAFT – DO NOT QUOTE
4. The upstairs was not receiving enough heat as observed by the homeowners. The
contractor installed two sheet metal scoops to affect airflow and heating.
Commissioning
Commissioning has its roots in the shipbuilding industry, and has made its way to the industrial
sector, but is not common practice for residences. Our experience with construction practices in
residences leads us to think that commissioning is a good idea when a house is newly
constructed, before or after a major remodel, or when ownership changes hands. We also found
that an important part of the commissioning process is to have detailed oversight of contractor
activities. Houses need maintenance, and problems can arise, unknown to the homeowner.
Wray et al. (2003) have developed a recommended commissioning procedure for residences,
which is similar to the diagnostics we have outlined in this paper. A flow chart of the
recommended commissioning procedure, along with references to protocols for each of the
testing procedures, can be found in Appendix D. A sample commissioning report can also be
found for this house in the pre-retrofit condition.
Diagnostics and Screening in Four Cold Climate Houses
A survey of four houses with basements were included in this study in order to see if the
checklist would apply differently to these houses where HVAC systems are often located in
these semi-conditioned basement spaces. These houses were in two cold climates: Minnesota
and Massachusetts, where basement construction is common. These houses were diagnosed, and
recommendations were made, but they were not retrofitted as part of the study. The summary of
diagnostic results, and recommendations for these houses can be found in Appendix G. All four
of the cold climate houses had full basements. We pressurized the houses with the basement
door closed and then measured the pressure difference between the basement and the house.
Three of the basements that we tested were essentially inside the air boundary of the house. The
other basement was about half inside and half outside. All the houses had at least one supply
register in the basement. One of the basements was insulated on the interior of the foundation
walls, one house had insulation on half of the basement ceiling and the band joist, but no
insulation on the foundation walls, an the other two houses had insulation on some of the
foundation walls, and some of the basement ceiling, but not in a consistent manner. Our
recommendation in all of these cases was to bring the basement inside the pressure boundary by
airsealing the rim joist and any penetrations between the basement and outside. We also
recommended to insulate the basement walls wherever they were not insulated. This brings the
basement ducts inside the envelope so there are no losses to outside. The basement ducts would
only need to be sealed if they are so significant that they are causing distribution problems. If
the basement cannot be brought inside the pressure boundary of the house (because the leaks
cannot be found and sealed) then the evaluation of whether to seal the ducts can be made using
the duct pressurization test or the delta q test, which both give an estimation of duct leakage to
outside.
One issue that needs to be addressed if the basement is brought into conditioned space is that the
combustion appliances are generally located in this space, and some provisions need to be made
38
DRAFT – DO NOT QUOTE
for makeup combustion air. The best option is to install sealed combustion appliances, which
take combustion air directly from outside. The appliances also have powered exhaust vents so
the danger of back drafting is eliminated. One of the houses that we surveyed had a sealed
combustion furnace and water heater. One of the houses had an outdoor air inlet into the return
duct, and a supply register (but no return) in the room where the furnace was located. Thus, the
furnace room is pressurized to avoid back drafting of the furnace and water heater. This practice
complies with Minnesota Building Code, but we don't recommend it because of the additional
heat load on the house due to the outdoor air inlet into the return.
Lessons Learned for Best Practices Guidelines
Our experience from the field pilot study has given us some insights for the best practices
guidelines (Walker 2003). For example, we had planned to retrofit the attic to be an insulated
unvented space, however the building code officials were not happy with this. Various options
were discussed – including creating a vented space behind a radiant barrier, adding windows to
make it “living space”, and the need to put drywall over the roof deck insulation because of the
fire hazard associated with having the furnace in the attic. With sufficient time and resources we
probably could have persuaded the code authorities to allow a sealed attic, however, because we
needed to complete the retrofits as soon a possible, we retained the vented attic and buried the
ducts in additional ceiling insulation. This method of burying the ducts also eliminates radiant
exchange between duct surfaces and hot undersides of the pitched roof surfaces during air
conditioning operation. The caveat is that burying ducts in insulation may not be suitable in
more humid climates where there is the potential of condensation on the outside of the ducts.
Like most people considering renovations, we had limits of time and budget, that meant that we
had to provide less innovative solutions. In other new construction projects we have worked on
we have faced the same issues – innovations in construction practice are strongly resisted by
building code officials. This resistance severely limits the range of renovations that can be easily
carried out. Given the strict conservatism of code officials, it is unlikely that these issues can be
dealt with on an individual project basis without extensive advance planning. Rather, we need to
be able to show specific examples of retrofits that are proposed. We hope that with research
projects like the current study, and the continuing efforts of other building scientists that
sufficient evidence for accepting some innovative building changes will become more widely
accepted.
We spent considerable time with a contractor looking at many retrofit options for the test house.
A key issue was where to locate the air handler. Including an economizer in the new system
meant that the existing garage location was not useable because of the difficulty in running the
large ducts associated with the economizer. The alternative was to put the air handler in the attic
(this is the standard retrofit performed by the contractor) and reduce the attic temperature
extremes. As discussed above – this created additional issues with code officials.
Other issues that need to be considered are the need to ensure good communication with the
homeowner and be able to discuss homeowner concerns (particularly if the retrofit is
innovative). It is essential to find a contractor who is willing to do something out of the ordinary
and persuade them to change the equipment they use and work crew practices. Ensuring that the
39
DRAFT – DO NOT QUOTE
work is a collaborative effort has benefits because contractors may have good ideas they want to
try out or, knowledge of alternative equipment. When having a contractor do something new or
unusual they need constant supervision to ensure that you get what you want (For example., we
had wanted to reuse the non-leaky low flow resistance sheet metal duct already installed in our
test house, but the installers removed it all and replaced it with flexible duct, which is what they
normally do in a retrofit situation).
Lessons Learned
1. Code officials are a significant barrier to the adoption of innovative retrofits.
2. Contractors, like code officials, are unfamiliar with innovative retrofits and need to be
willing to learn on the job. Innovative retrofits have higher labor costs (and possibly
material as well) until the contractor has gained enough experience to adequately finish a
job with fewer hours.
3. Ensure that the work is a collaborative effort – contractors may have good ideas they
want to experiment with or knowledge of alternative equipment that can be valuable.
4. Constant communication and supervision is required to ensure extraordinary retrofits are
carried out as desired. A pre-agreed upon workplan between the contractor and client
would aid in the communication process.
5. Develop a manual and checklist to help installers.
6. Keep thorough notes and communicate this information with installers.
7. Installers need training on new equipment/techniques.
8. Commissioning of retrofitted systems is required, and more complex systems like zoning
dampers and variable speed systems need to be thoroughly checked for proper design
operation.
40
DRAFT – DO NOT QUOTE
Acknowledgements
This work was supported by and the Assistant Secretary for Energy Efficiency and Renewable
Energy, Building Technologies, of the US Department of Energy (DOE) under contract No. DE-
AC03-76SF00098. The authors would like to thank Darryl Dickerhoff, Duo Wang, Douglas
Brenner, Brian Smith and Nance Matson of LBNL, Ananda Harzell (CSG), Rick Wylie (Beutler
Heating and Air Conditioning), and Stacy Hunt (IBACOS).
41
DRAFT – DO NOT QUOTE
References
Abushakra, B., I. Walker and M. Sherman. 2002. “A Study of Pressure Losses in Residential
Air Distribution Systems”. Proceedings of ACEEE Summer Study 2002, American Council for
an Energy Efficient Economy. Washington, D.C. Lawrence Berkeley National Laboratory
Report LBNL 49700. Berkeley, CA.
ASTM E779-99. 1999. “Standard Test Method for Determining Air Leakage Rate by Fan
Pressurization” Annual Book of ASTM Standards Vol. 04.11. American Society for Testing
and Materials.
ASTM E1554-94. 1994. “Standard Test Methods for Determining External Air Leakage of Air
Distribution Systems by Fan Pressurization”. Annual Book of ASTM Standards Vol. 04.11.
American Society for Testing and Materials.
ASTM E1186-03. 2003. "Standard Practices for Air Leakage Site Detection in Building
Envelopes and Air Retarder Systems". Annual Book of ASTM Standards Vol. 04.11. American
Society for Testing and Materials.
ASTM E1827-96. 2002. "Standard Test Methods for Determining Airtightness of Buildings
Using an Orifice Blower Door". Annual Book of ASTM Standards Vol. 04.12. American
Society for Testing and Materials.
Cool Roofing Materials Database (CRMD). 2001. http://eetd.lbl.gov/
coolroof. Lawrence
Berkeley National Laboratory. Berkeley, CA.
Consortium for Energy Efficiency (CEE). 2000. “Specification of Energy-Efficient Installation
and Maintenance Practices for Residential HVAC Systems”. Boston, MA.
Jump, D. and M. Modera. 1994. “Energy Impacts of Attic Duct Retrofits in Sacramento
Houses”. Lawrence Berkeley National Laboratory Report LBNL-35375. Berkeley, CA.
Jump, D., I. Walker and M. Modera. 1996. “Field Measurements of Efficiency and Duct
Effectiveness in Residential Forced Air Distributions Systems”. Lawrence Berkeley National
Laboratory Report LBNL-38537. Berkeley, CA.
Sherman, M. and D. Dickerhoff. 1994. "Air-Tightness of U.S. Dwellings,'' In Proceedings, 15th
AIVC Conference: The Role of Ventilation, Vol. 1, Coventry, Great Britain:Air Infiltration and
Ventilation Centre, 1994, pp. 225-234.
Walker, I., D. Dickerhoff and M. Sherman. 2002. “The DeltaQ Method of Testing the Air
Leakage of Ducts”. Proceedings of ACEEE Summer Study 2002, American Council for an
Energy Efficient Economy. Washington, D.C. Lawrence Berkeley National Laboratory Report
LBNL 49749. Berkeley, CA.
42
DRAFT – DO NOT QUOTE
Walker, I. and M. Modera. 1998. “Field Measurements of Interactions Between Furnaces and
Forced Air Distributions Systems”. ASHRAE Transactions Vol. 104, Pt. 1B, pp.1805-1816
ASHRAE. Atlanta, GA.
Walker, I., M. Sherman, M. Modera and J. Siegel. 1998. “Leakage Diagnostics, Sealant
Longevity, Sizing and Technology Transfer in Residential Thermal Distribution Systems”.
Lawrence Berkeley National Laboratory Report LBNL-41118. Berkeley, CA.
Walker, I., M. Sherman, J. Siegel and M. Modera. 1999. “Effects of Duct Improvement and
EnergyStar Equipment on Comfort and Energy Efficiency”. Lawrence Berkeley National
Laboratory Report LBNL-43723. Berkeley, CA.
Walker, I., M. Sherman, J. Wempen, D. Wang, J. McWilliams and D. Dickerhoff. 2001.
“Development of a New Duct Leakage Test: DeltaQ". Lawrence Berkeley National Laboratory
Report LBNL 47308. Berkeley, CA.
Walker, I., C. Wray, D. Dickerhoff and M. Sherman. 2002. “Evaluation of Flow Hood
measurements for Residential Register Flows”. Lawrence Berkeley National Laboratory Report
LBNL 47382. Berkeley, CA.
Walker, I.S. and Wray, C.P. 2003. "Evaluation of flow capture techniques for measuring HVAC
grille air flows". ASHRAE Transactions (in p ress) Lawrence Berkeley National Laboratory
Report LBNL-51550. Berkeley, CA.
Walker, I.S. 2003. Best Practices Guide for Residential HVAC Retrofits. LBNL 53592.
Walker, I.S. and Sherman, M.H.. 2003. “Sealant Longevity for Residential Ducts”. Durability of
Building and Construction Sealants and Adhesives, ASTM STP 1453, A. Wolf Ed., American
Society for Testing and Materials, West Conshohocken, Pa. (LBNL 50189).
Wray, C. and M. Sherman. 2001. "Residential Commissioning to Assess Envelope and HVAC
System Performance". Lawrence Berkeley National Laboratory Report LBNL-47412. Berkeley,
CA. http://epb.lbl.gov/Publications/lbnl-47412.pdf.
Wray, C., I. Walker and M. Sherman. 2002. “Accuracy of Flowhoods in Residential
Applications”. Proceedings of ACEEE Summer Study 2002, American Council for an Energy
Efficient Economy. Washington, D.C. Lawrence Berkeley National Laboratory Report LBNL
49697. Berkeley, CA.
Wray, C., I. Walker, J. Siegel and M. Sherman. 2002. "Practical Diagnostics for Evaluating
Residential Commissioning Metrics". Lawrence Berkeley National Laboratory Report LBNL-
45959. Berkeley, CA. http://epb.lbl.gov/Publications/lbnl-45959.pdf
.
43
DRAFT – DO NOT QUOTE
Wray, C., I. Walker and M. Sherman. 2003. "Instrumented Home Energy Rating and
Commissioning". Lawrence Berkeley National Laboratory Report LBNL-52216. Berkeley, CA.
http://epb.lbl.gov/Publications/lbnl-52216.pdf.
US Department of Energy (USDOE). 2001. “Choosing or Upgrading Your Central Air
Conditioner”. Office of Building Technology, State and Community Programs”.
http://www.eren.doe.gov/buildings/heatcool_cenair.html.
44
DRAFT – DO NOT QUOTE
Appendix A. Field Surveys of Four Houses in California for Retrofitting
Table A.1. Diagnostic Checklist for Concord House.
Measurement/
Observation
Potential Target value Actual Value Potential Retrofit
Action
Duct leakage <10% of air handler flow S 12%; R 33% Seal ducts: aeroseal/tape/mastic
Duct insulation R6 (RSI 1) to R8 (RSI 1.4) for all ducts
outside conditioned space
~R2 Add insulation to ducts
Air flows at registers Compare to ACCA manual J Sum of supplies and
returns: S 683; R 540.
Several duct runs had
low airflow.
Replace registers, open/close
dampers.
Air handler flow
Cooling: 1400 based on nameplate, 900 based
on measured capacity.
Heating: 12.5 cfm/kBtu/h
804 cfm Replace filters, fix duct restrictions,
change fan speed, replace fan with
high efficient unit, add extra returns
in return restricted systems
Filter Condition Clean and at least MERV 6
5
Non-operational
electrostatic
Replace with MERV 6 or better. Use
4 inch filters.
Thermostat Setting
Heating: 68°F (20°C) Cooling: 78°F (25°C)
Programmable None
Spot ventilation 50 cfm each bathroom
100 cfm each kitchen
Master bath 76 cfm; up
bath 63 cfm; down bath
28 cfm; kitchen 238
cfm
None
Spot Ventilation fan
power consumption
2.5 cfm/W (1.2 L/s/W). Look up in HVI
directory (www.hvi.org
)
N/A None
Equipment capacity Manual S: 3.5 tons on nameplate 2.3 tons measured Replace with more efficient and
slightly smaller unit.
Refrigerant charge Use superheat or subcooling tests Leak in refrigerant line Fix leak and add refrigerant or
replace unit.
Age and Condition of
HVAC system
Clean and undamaged.
Determine system age.
Poor condition. Age
unknown.
Replace system.
Location of HVAC
system equipment and
ducts
Inside conditioned space
Garage Seal and insulates duct locations to
make them more like conditioned
space, or move system location.
Window A/C units
EnergyStar compliant
None None
Multiple systems/zoning System and controls in good working order
and providing good comfort for occupants
None Upgrade to zoned system to address
stratification between the first and
second floors.
Envelope leakage Normalized Leakage Area reduction of 0.35 ELA: 278 in2; NLA
0.95
Airseal plumbing penetrations, attic
chases, top plates, and other
unintentional openings. Reduce to
NLA 0.6
Moisture testing No moisture problems None None
House insulation
Ceiling: R-30 (RSI 5.3) minimum, R-49 (RSI 8.6)
in cold/severe cold climate. Floor over
crawlspace:R- 25 (RSI 4.4). Basement walls: R10
(RSI 1.8), Basement Floor or slab usually
depends on local codes. Walls: Cavity should be
completely filled with insulation.
Attic: 4-5 inches
cellulose (~R 15).
Walls: R-11
Add at least 6" cellulose insulation to
attic.
Windows Double-glazed, low-e. Shaded in cooling
dominant climates
Single glazed aluminum Replace windows. (Alarm system
must be replaced at same time.).
Window shading
Located on south and/or west facing windows
None Add shading to reduce solar loads
Solar radiation control Radiant barrier in attic, low absorptivity roof
coatings
None Add reflective paint to roof and/or
sunshades for windows.
Wall, floor and ceiling
construction
Space for ducts/ vents/ insulation Attic has plenty of
space.
Add insulation in attic.
Occupant survey
Ask occupants to report
problems
No problems Upstairs hotter than
downstairs. Feels dust
blowing when AH is
on. Windows rattle.
Add zoned system to condition first
and second floor independently.
Duct sealing for dust being drawn in
through return. Possibly replace
windows.
5
MERV is an industry standard rating system for air filters, it stands for Minimum Efficiency
Report Value determined using ASHRAE Standard 52.2
45
DRAFT – DO NOT QUOTE
Table A.2. Diagnostic Checklist for Larkspur House.
Measurement/
Observation
Potential Target value Actual Value Potential Retrofit
Action
Duct leakage <10% of air handler flow S 10%; R 17% Seal ducts: aeroseal/tape/mastic
Duct insulation R6 (RSI 1) to R8 (RSI 1.4) for all ducts outside
conditioned space
? Add insulation to ducts
Air flows at registers Compare to ACCA manual J Sum of supplies and
returns: S 1120; R
N/A.
Replace registers, open/close
dampers.
Air handler flow
Cooling: 1200 cfm based on nameplate
Heating: 12.5 cfm/kBtu/h
1216 cfm None
Filter Condition Clean and at least MERV 6
6
Poor fitting air filter Replace with new MERV 6 or
better.
Thermostat Setting
Heating: 68°F (20°C) Cooling: 78°F (25°C)
Programmable None
Spot ventilation 50 cfm each bathroom
100 cfm each kitchen
N/A None
Spot Ventilation fan
power consumption
2.5 cfm/W (1.2 L/s/W). Look up in HVI directory
(www.hvi.org
)
N/A None
Equipment capacity Manual S: 3 tons on nameplate 2.8 tons measured None
Refrigerant charge Use superheat or subcooling tests Charged in 2000 None
Age and Condition of
HVAC system
Clean and undamaged.
Determine system age.
Condensing coil
dirty. System 27
years old.
Clean coil or replace system.
Location of HVAC system
equipment and ducts
Inside conditioned space
Crawlspace Seal and insulates duct locations
to make them more like
conditioned space, or move system
location.
Window A/C units
EnergyStar compliant
None None
Multiple systems/zoning System and controls in good working order and
providing good comfort for occupants
None None
Envelope leakage Normalized Leakage Area reduction of 0.35 ELA: 340 in2; NLA:
1.27
Airseal plumbing penetrations,
ceiling tongue and groove, and
other unintentional openings.
Reduce to NLA 0.93
Moisture testing No moisture problems None None
House insulation
Ceiling: R-30 (RSI 5.3) minimum, R-49 (RSI 8.6) in
cold/severe cold climate. Floor over crawlspace:R- 25
(RSI 4.4). Basement walls: R10 (RSI 1.8), Basement
Floor or slab usually depends on local codes. Walls:
Cavity should be completely filled with insulation.
Attic: unknown - no
access. Walls: R-11.
Crawlspace: R-11.
Upgrade crawlspace insulation to
R 25.
Windows Double-glazed, low-e. Shaded in cooling dominant
climates
Single pane
aluminum frame
Replace with double pane low e.
Window shading
Located on south and/or west facing windows
None Add shading to reduce solar loads
Solar radiation control Radiant barrier in attic, low absorptivity roof
coatings
None Add reflective paint to roof and/or
sunshades for windows.
Wall, floor and ceiling
construction
Space for ducts/ vents/ insulation Crawlspace has
plenty of space.
Add insulation in crawlspace.
Occupant survey
Ask occupants to report
problems
No problems No problems noted None.
6
MERV is an industry standard rating system for air filters, it stands for Minimum Efficiency
Report Value determined using ASHRAE Standard 52.2
46
DRAFT – DO NOT QUOTE
Table A.3. Diagnostic Checklist for Moraga House.
Measurement/
Observation
Potential Target value Actual Value Potential Retrofit Action
Duct leakage <10% of air handler flow Two systems: S1 22%;
R1 14% S2 10% R2 N/A
Seal ducts in system 1:
aeroseal/tape/mastic
Duct insulation R6 (RSI 1) to R8 (RSI 1.4) for all ducts
outside conditioned space
R4 Add insulation to ducts
Air flows at registers Compare to ACCA manual J Sum of supplies and
returns: S1 760; R1 834;
S2 486, R2 585. Low
flows in one bedroom and
two bathrooms.
Replace registers, open/close dampers.
Air handler flow
Cooling: System 1:1600; System 2: 1000.
Heating: 12.5 cfm/kBtu/h
System 1: 970 cfm
System 2: 540 cfm
Replace filters, fix duct restrictions,
change fan speed, replace fan with high
efficient unit. Add a second return in
system 2, remove two supplies from
system 1 and add to system 2.
Filter Condition Clean and at least MERV 6 Dirty filter Replace with MERV 6 or better.
Thermostat Setting
Heating: 68°F (20°C) Cooling: 78°F
(25°C)
Programmable None
Spot ventilation 50 cfm each bathroom
100 cfm each kitchen
Bathrooms: 75 cfm; 63
cfm; 28 cfm. Kitchen 240
cfm.
None
Spot Ventilation fan
power consumption
2.5 cfm/W (1.2 L/s/W). Look up in HVI
directory (www.hvi.org
)
N/A None
Equipment capacity Manual S 4 tons on system 1
nameplate. 2.5 tons on
system 2 nameplate.
None
Refrigerant charge Use superheat or subcooling tests System 1: Actual/6
Target/13
System 2: Actual/6
Target/11
Both systems are
overcharged.
Remove refrigerant
Age and Condition of
HVAC system
Clean and undamaged.
Determine system age.
System 1: 16 yrs old
System 2: 7 yrs old - good
condition
Replace system 1.
Location of HVAC
system equipment and
ducts
Inside conditioned space
Both systems in closets. None
Window A/C units
EnergyStar compliant
None None
Multiple systems/zoning System and controls in good working order
and providing good comfort for occupants
Two systems serving
opposite ends of house.
None
Envelope leakage Normalized Leakage Area reduction of
0.35
ELA: 350 in2; NLA 0.69 Airseal plumbing penetrations, attic
chases, top plates, and other
unintentional openings. Reduce to
NLA 0.34
Moisture testing No moisture problems None None
House insulation
Ceiling: R-30 (RSI 5.3) minimum, R-49 (RSI
8.6) in cold/severe cold climate. Floor over
crawlspace:R- 25 (RSI 4.4). Basement walls:
R10 (RSI 1.8), Basement Floor or slab
usually depends on local codes. Walls: Cavity
should be completely filled with insulation.
Attic: 5 inches cellulose
(R 17). Walls: R-11.
Crawlspace: probably
some in the new part.
Add at least 6" cellulose insulation to
attic. Add R-25 insulation to old
crawlspace if none existing.
Windows Double-glazed, low-e. Shaded in cooling
dominant climates
Double glazed. None
Window shading
Located on south and/or west facing
windows
Shielded by trees
south/west
None
Solar radiation control Radiant barrier in attic, low absorptivity
roof coatings
None Add reflective paint to roof and/or
sunshades for windows.
Wall, floor and ceiling
construction
Space for ducts/ vents/ insulation Attic and crawlspace have
space.
Add insulation in attic. Add insulation
in crawlspace if needed.
Occupant survey
Ask occupants to report
problems
No problems Hard to cool the family
room.
Add ducts to system 2 to go to family
room registers and disconnect registers
from system 1 to family room.
47
DRAFT – DO NOT QUOTE
Table A.4. Diagnostic Checklist for Castro Valley House.
Measurement/
Observation
Potential Target value Actual Value Potential Retrofit
Action
Duct leakage <10% of air handler flow S 9%; R 5% Seal ducts: aeroseal/tape/mastic
Duct insulation R6 (RSI 1) to R8 (RSI 1.4) for all ducts outside
conditioned space
R 4 Add insulation to ducts
Air flows at registers Compare to ACCA manual J Sum of supplies and
returns: S 823; R 940.
Replace registers, open/close
dampers.
Air handler flow
Cooling: 1270 based on nameplate.
Heating: 12.5 cfm/kBtu/h
1160 cfm Replace filters, fix duct restrictions,
change fan speed.
Filter Condition Clean and at least MERV 6 ? Replace with MERV 6 or better.
Thermostat Setting
Heating: 68°F (20°C) Cooling: 78°F (25°C)
Programmable None
Spot ventilation 50 cfm each bathroom
100 cfm each kitchen
Bathrooms: 280 cfm; 60
cfm; 30 cfm; Kitchen
700 cfm
Check that kitchen vent fan does
not cause furnace backdraft
potential.
Spot Ventilation fan
power consumption
2.5 cfm/W (1.2 L/s/W). Look up in HVI
directory (www.hvi.org
)
N/A None
Equipment capacity Manual S: 3.6 tons on nameplate None
Refrigerant charge Use superheat or subcooling tests Actual superheat was 24
Target superheat was
11. System is severely
undercharged
Add refrigerant.
Age and Condition of
HVAC system
Clean and undamaged.
Determine system age.
Good condition. 10
years old.
None
Location of HVAC
system equipment and
ducts
Inside conditioned space
Closet None
Window A/C units
EnergyStar compliant
None None
Multiple systems/zoning System and controls in good working order and
providing good comfort for occupants
None None
Envelope leakage Normalized Leakage Area reduction of 0.35 ELA: 269 in2; NLA
1.06
Airseal plumbing penetrations, attic
chases, top plates, kneewall areas,
and other unintentional openings.
Reduce below NLA 0.71
Moisture testing No moisture problems None None
House insulation
Ceiling: R-30 (RSI 5.3) minimum, R-49 (RSI 8.6)
in cold/severe cold climate. Floor over
crawlspace:R- 25 (RSI 4.4). Basement walls: R10
(RSI 1.8), Basement Floor or slab usually depends
on local codes. Walls: Cavity should be completely
filled with insulation.
Attic: 5 inches foam (R
35). Walls: R-19.
Crawlspace: fiberglass.
None
Windows Double-glazed, low-e. Shaded in cooling
dominant climates
Double glazed low e None
Window shading
Located on south and/or west facing windows
Shaded by garage and
trees on the south and
east.
None
Solar radiation control Radiant barrier in attic, low absorptivity roof
coatings
None Add reflective paint to roof and/or
sunshades for windows.
Wall, floor and ceiling
construction
Space for ducts/ vents/ insulation Crawlspace has space
for additional ducts.
None
Occupant survey
Ask occupants to report
problems
No problems Too hot in the master
bedroom when heating,
undercooled office
when cooling.
Improve air flows into affected
rooms – reduce airflow into master
bedroom – add window shading to
office.
48
DRAFT – DO NOT QUOTE
Appendix B: Long-Term Monitoring Results
The Concord test house was monitored during the pre-retrofit period for about eight weeks from
12 July through 18 September 2002. The house envelope and HVAC system were then
retrofitted. Data collection then resumed on 2 October for the post-retrofit period.
Instrumentation and data acquisition for continuous monitoring of the on-site weather conditions,
HVAC system component power, and supply, return, zonal and attic conditions were installed at
the test house. The continuous monitoring used a computerized data logging system (Photo B.1)
to collect 24 channels of data at 10-second intervals, which were downloaded each evening at
midnight via a telephone connection to a computer at LBNL. The monitored parameters and
instrumentation used are displayed in Table B.1.
The analysis was conducted using a data domain including days only when the air conditioner
was in operation, which was 34 non-continuous days for the pre-retrofit period (any
discontinuities between days in the presentation of the monitored data are due to the non-
continuous data domain). All data are presented in Pacific Standard Time (PST).
Due to several delays during this project we were unable to complete cooling season data post-
retrofit for inclusion in this report. However, we are continuing the monitoring and hope to
perform a comparison between pre and post retrofit performance at a later date.
49
DRAFT – DO NOT QUOTE
Table B.1. Monitored parameters and instrumentation.
Weather Conditions (5) Instrumentation
drybulb temperature aspirated thermocouple
relative humidity aspirated capacitive humidity sensor
insolation (total incoming solar radiation) silicon photodiode pyranometer
wind speed 3-cup anemometer
wind direction wind vane
HVAC System Components (4)
compressor motor power meter / current transformer
condenser fan motor power meter / current transformer
air-handler fan motor power meter / current transformer
Refrigerant Conditions (4)
suction temperature thermocouple
discharge temperature thermocouple
suction pressure pressure transducer
discharge pressure pressure transducer
Supply Air Conditions (4)
temperature (single sensor after heat exchanger) thermocouple
temperature (before heat exchanger/coil) thermocouple
relative humidity capacitive humidity sensor
temperature (register) thermocouple
Return Air Conditions (3)
temperature (downstairs) thermocouple
temperature (plenum) thermocouple
relative humidity capacitive humidity sensor
Zone Air Conditions (2)
temperature (downstairs) thermocouple
temperature (upstairs) thermocouple
Attic Air Conditions (2)
temperature thermocouple (aspirated)
relative humidity capacitive humidity sensor (aspirated)
what do up, down, plenum mean above?
Monitored On-Site Weather Data
The on-site weather data that were monitored include outdoor drybulb temperature and outdoor
relative humidity in an aspirated tube under the roof eave (Photo B.2), total incoming solar
radiation (insolation) on the roof gutter (Photo B.3), and wind speed and direction atop a fifteen
foot high tower (Photo B.4). These data with the exception of wind direction and the inclusion
of the calculated outdoor enthalpy are shown in Figure B.1. The outdoor temperature ranged
from 12 to 40°C (54 to 104°F), outdoor relative humidity from 10-80%, outdoor enthalpy from
18-34Btu/lbm (kJ/kg), insolation peaked at 775W/m
2
and wind speed peaked at 6mph.
50
DRAFT – DO NOT QUOTE
Photo B1. Data acquisition equipment in
garage with existing furnace in background.
Photo B.2. Outdoor air aspirated
temperature and relative humidity sensors
located under roof eave.
Photo B.3. Location of
pyranometers on roof with
wind tower in background
Photo B.3. Pyranometers
located on roof platform
[left diffuse and right
total].
.
Photo B.4. Wind tower with
speed and direction sensors.
51
DRAFT – DO NOT QUOTE
Outdoor Drybulb Temperature [C]
10
15
20
25
30
35
40
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Humidity Ratio
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
0.011
0.012
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Outdoor Enthalpy [Btu/lbm]
0
10
20
30
40
50
60
70
80
12-Jul 22-Jul 1-Aug 11-Aug 21-Aug 31-Aug 10-Sep
Figure B.1. Monitored 10-second on-site weather data for the 34 day pre-retrofit
period [drybulb temperature, humidity ratio, and enthalpy].
52
DRAFT – DO NOT QUOTE
Monitored HVAC System Data
The HVAC system data that were monitored include compressor and condenser fan
motors (Photo B.5), and the air-handler motor (Photo B.6). These data with the
exception of vent position and the inclusion of the total air-conditioner power are shown
in Figure B.2. The total AC power peaked at about 7kW, with the air-handler peaking at
about 0.55kW, the condenser fan at about 0.45kW and the compressor at about 6kW.
Photo B.5. AC
condensing unit
power meters, one
each for the
compressor and
condensing fan
motors.
Photo B.6. Air handler power meter
located in garage.
Photo B.7. Attic air aspirated
temperature and relative humidity
sensors located near attic ceiling.
53
DRAFT – DO NOT QUOTE
Compresor Power [W]
0
1000
2000
3000
4000
5000
6000
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Condenser Fan Power [W]
0
50
100
150
200
250
300
350
400
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Air-Handler Power [W]
0
50
100
150
200
250
300
350
400
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Total Air Conditioning Power [W]
0
1000
2000
3000
4000
5000
6000
7000
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Figure B.2. Monitored 10-second air-conditioning component electricity demand data [W]
54
DRAFT – DO NOT QUOTE
Monitored Refrigerant Data
The refrigerant data are discussed in the Performance of Pre-Retrofit Air-Conditioner section in
this chapter and in the Refrigerant Charge section of the Diagnostics chapter.
Monitored Supply and Return Air Data
The supply and return air data that were monitored include plenum takeoffs for upstairs and
downstairs drybulb temperature and downstairs relative humidity (Figure 16). These data are
shown in Figure B.3. The upstairs supply air temperature operated at about 15°C, the
downstairs temperature and humidity at about 12°C and 70%, the downstairs return air
temperature operated at about 26°C, the return plenum temperature and humidity at about 25°C
and from 25-50%.
55
DRAFT – DO NOT QUOTE
Supply Air Temperature [C]
10
15
20
25
30
35
40
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Supply Air Relative Humidity
10
20
30
40
50
60
70
80
90
100
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Return Air Temperature [C]
10
15
20
25
30
35
40
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Return Air Relative Humidity
10
20
30
40
50
60
70
80
90
100
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Figure B.3. 10-second supply and return data in the pre-retrofit period.
56
DRAFT – DO NOT QUOTE
Monitored Zonal and Attic Air Data
The zonal and attic air data that were monitored include downstairs thermostat temperature, attic
temperature and attic relative humidity (Photo B.7). These data are shown in Figure B.1(e).
The downstairs thermostat temperature was set at 25.5°C (78°F) and normally was between 23
and 26°C. The attic air temperature ranged from 10°C nighttime to 58°C daytime, and humidity
ranged from 18 to 48%.
Thermostat Temperature [C]
10
15
20
25
30
35
40
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Attic Air Temperature [C]
10
20
30
40
50
60
70
80
90
100
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Attic Air Relative Humidity
10
20
30
40
50
60
70
80
90
100
12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep
Figure B.1(e). Monitored 10-second zonal air temperatures, and attic air temperature and
relative humidity data in the pre-retrofit period.
57
DRAFT – DO NOT QUOTE
Performance of Pre-Retrofit Air-Conditioner
The performance of the pre-retrofit air-conditioner during the 4 hottest days of the pre-retrofit
period (days 18,19, 26 & 27) is presented in this section. It includes an analysis of AC
component (i.e., compressor motor, condenser fan motor and air-handler motor) power demand,
calculation of operating efficiency and cooling capacity, cycling behavior and superheat.
During this 4-day period the outdoor air temperature peaked at 37.5-39.5 °C (99.5-103 °F) and
the AC electricity consumption was highest (37.6 to 46.9 daily kWh). On average for these 4
days the air-conditioner peaked at 6.6 kW and ran steadily at 6.0 kW during the hottest time of
the day and at 5.0 kW during the cool evening hours. The transient start-up power was 500-600
W above steady-state operation. The AC typically drew about 5.5 kW.
The compressor motor peaked on cycle start-up at 5.8 kW and settled into steady-state operation
at 5.4 kW during the daytime and 4.2 kW during the evening. The transient start-up power was
400 W above steady-state operation. On average the compressor drew about 4.8 kW.
The condenser fan motor occasionally peaked on cycle start-up at 400-450 W and settled into
steady-state operation at 360 W during the daytime and 380 W during the evening. The evening
time operation was 20 W higher than daytime because of a cooler outdoor air temperature and
greater air density. The transient start-up power was 50-100 W above steady-state operation.
On average the condenser fan drew about 370 W.
The air-handler motor peaked on cycle start-up in about one-half of the cycles at 450-500 W and
settled into steady-state operation at 320-340 W. The transient start-up power was 40-160 W
above steady-state operation. On average the air-handler drew about 330 W.
The operating AC energy efficiency ratio (EER) ranged from 4.2-6.0 on days 18 & 19 and 3.6-
5.8 on days 26 & 27. The operating cooling capacity ranged from 25-30 kBtu/h on days 18 & 19
and 21-28 kBtu/h on days 26 & 27. The cooling coil temperature drop ranged from 12-14 °C on
days 18 & 19 and 10-13 °C on days 26 & 27. The low-end EER and capacity occurred during
the hot part of the day (11am-4pm) and both increased with decreasing outdoor air temperature
as the evening progressed (discussed in detail in the Dependence of AC Performance on Outdoor
Air Temperature section of this chapter). The calculation method for operating EER, cooling
capacity and coil temperature drop are described in Appendix C.
During this 4-day period the AC ran from about 11am-11pm (PST) and cycled on average 24.5
times. The air-conditioner exhibited a brief transient period of 50 seconds on start-up of each
cycle exceeding the steady-state operating power consumption. The condenser fan and air-
handler reached steady-state in 20 seconds. In an example transient the compressor peaked at
5.8 kW and ran steady at 5.2 kW ( = 600 W), the condenser fan peaked at 410 W and ran
steadily at 360 W ( = 50 W), the air-handler peaked at 470 W and ran steadily at 320 W ( =
150 W), and the whole a/c peaked at 6.6 kW and ran steadily at 5.8 kW ( = 800 W). For this
example the operating EER, cooling capacity and coil temperature drop were 4.3, 25 kBtu/h and
12.5°C.
58
DRAFT – DO NOT QUOTE
The air-conditioner superheat was calculated from a steady cycle on the hottest day with these
parameters: outdoor drybulb temperature [37°C], return plenum air temperature [25°C], return
plenum relative humidity [41%], refrigerant suction pressure [120 psi] and refrigerant suction
temperature [30°C]. The desired and actual superheat were 0°C and 16°C for the R-22
refrigerant, calculated as described in the Refrigerant Charge section of the Diagnostics chapter.
The 10-second data for the 34 non-continuous days were averaged (totaled for air-conditioning
data) into hourly data sets [equations 1 and 2]. The units for total hourly AC electricity use are
in [Wh].
=
=
=
360
1
,
1
_
360
i
jijhouravg
XX Eq. 1
=
=
=
360
1
,
1
_
10360
i
jijhourtotal
sWWh Eq. 2
The results show that AC use peaked at 5pm PST lagging behind the outdoor air temperature by
one hour and peak insolation lagged behind the outdoor air temperature by three hours.
Dependence of Air-Conditioner Performance on Outdoor Air Temperature
A statistical analysis of the dependence of air-conditioning performance on outdoor air
temperature was completed. Also, the relationship between attic air temperature and outdoor air
temperature is examined in this section. The analysis domain included only the 4 hottest days
for the 10-second data and all 34 days for the daily data.
The 10-second data for the 34 non-continuous days were averaged for non-AC parameters
[equation 3] and the hourly AC data were totaled [equation 4] into daily data sets. The units for
total daily AC electricity use are in [kWh].
=
=
8640
1
1
_
8640
i
idailyavg
XX Eq. 3
=
=
24
1
1
_
1000
j
jdailytotal
WhkWh Eq. 4
The statistical analysis firmly concluded that the outdoor drybulb air temperature was the
strongest predictor of AC component electricity demand and daily consumption, as well as
operating EER and cooling capacity. The single-variable regression statistics (analysis of
variance, coefficients and percent standard error) for the linear relationship in equation 5 are
detailed in
Table B.2.
ETEMPERATUROUTDOORCCnceACPerforma _*
10
+= Eq. 5
59
DRAFT – DO NOT QUOTE
It is noticeable in these data that there are two air handler power consumption levels
corresponding to two different air handler flows. We noticed that the jump in air handler power
corresponded to a date on which we had been at the house. On that day we had noticed that the
air handler cover panel was not properly installed, leaving a large gap in the cabinet for return
leakage at a high pressure. We took care to install the panel correctly with no gap when we
finished our work that day. The presence of absence of that gap could have caused a change in
air handler airflow.
Examination of the 10-second data shows the total AC power increased with increasing outdoor
air temperature at 69W/°C and likewise with compressor motor power at 70W/°C. However, the
condenser and air-handler fan power decreased with increasing outdoor air temperature at –1.2 &
-0.3W/°C. The Energy Efficiency Ratio (EER) decreased at -0.1/°C of outdoor air temperature
and cooling capacity decreased at -0.3kBtu/h/°C.
The mean AC power demand in this 4-day period was 5590W, where the component breakdown
was compressor motor 4900W, condenser fan motor 360W and air-handler fan motor 330W.
The mean EER and cooling capacity were 5 and 25.3kBtu/h.
Examination of the daily data reveals that total daily AC energy use increased at 4.7kWh/°C of
average daily outdoor air temperature. Average daily attic air temperature increased at 0.9°C/°C.
60
DRAFT – DO NOT QUOTE
Table B.2. Single-variable regression analysis of variance, coefficients and percent standard
error of 10-second and daily air-conditioning performance versus outdoor air temperature,
also, attic air temperature versus outdoor air temperature.
Data Limits Analysis of Variance Coefficients % Error
Dependent
Variable
lower upper n mean
σ
R
2
C
0
C
1
C
0
C
1
10-Second
AC 4500 6500 10783 5587 131 0.85 3259 68.5 0.3 0.4
Compressor 3800 6000 10789 4899 130 0.85 2516 70.1 0.4 0.4
Cond Fan 340 390 10756 358 3.9 0.68 400 -1.2 0.1 0.7
AH Fan 300 360 10762 330 8.4 0.03 342 -0.3 0.2 5.3
Capacity 20 32 11059 25.3 1.9 0.33 35 -0.3 0.4 1.4
EER 3 7 10424 5 0.4 0.68 9 -0.1 0.3 0.7
Attic Air 10 50 34560 32 2.6 0.92 0.86 1.2 5.8 0.2
Daily
AC 0 100 34 15 5.3 0.84 -89 4.7 9.0 7.6
Attic Air 0 100 34 28 0.5 0.95 9 0.9 9.0 4.1
Note: Prob>f and prob>T statistics for C
0
and C
1
were all 0.0001.
n: Number of observations.
prob>T. The probability of a greater absolute value for this T value.
prob>f: The probability of getting a greater F statistic than that observed if the hypothesis is
true. This is the significance probability.
σ: Root Mean Square Error is an estimate of the standard deviation of the error term. It is
calculated as the square root of the mean square error.
R
2
: Is a measure between 0 and 1 that indicates the portion of the total variation that is attributed
to the fit rather than left to residual error. It is also called the coefficient of determination and is
the square of the correlation between the dependent variable and the predicted values.
standard error (SE): Standard deviation of the parameter estimate.
61
Appendix C. Calculation of Operating Energy Efficiency Ratio (EER) and
Cooling Capacity
q
total
= q
sensible
+ q
latent
q
total
= ρ Q h
total
h
total
= h
return
- h
supply
q
sensible
= ρ Q h
sensible
h
sensible
= C
p
T
T = T
return
- T
supply
q
latent
= q
total
- q
sensible
q
latent
= ρ Q ( h
total
- h
sensible
)
q
latent
= ρ Q ( h
total
– C
p
T )
SHR = q
sensible
/ q
total
EER = cooling capacity [kBtu/h] / electrical input [kW]
COP = EER / 3.412
ρ
density of air lbm/ft
3
0.75
C
p
specific heat of air
Btu/lbm °F
0.24
Q air flow rate cfm 800
T
air temperature drop
°F
25
h
total
enthalpy change Btu/lbm 8.5
q
sensible
sensible cooling
capacity
Btu/h 21600
q
latent
latent cooling capacity Btu/h 9000
q
total
total cooling capacity Btu/h 30600
SHR sensible heating ratio dimensionless 0.71
electrical input electrical input kW 6000
EER energy efficiency ratio kBtu/h kW 5.1
COP coefficient of
performance
dimensionless 1.5
62
Appendix D. Residential Commissioning Procedures
NERAL RESIDENTIAL COMMISSIONING
AIR
DISTRIBUTION
SYSTEM?
3
8
START
1
2
56
7 10
KNOW
REFRIGERANT
CAPACITY
SPEC
?
TXV
?
12
13
14
COMBUSTION
APPLIANCES
IN CONDITIONED
SPACE
?
15 16
GENERATE
REPORT
DONE
Y
N
Y
N
Y
N
Y
N
SUGGESTED FLOWCHART: GE
ADDITIONAL
CIRCUMSTANCES
OR EQUIPMENT
TO TEST
?
4
11 9
OTHER
TESTS
Y
N
NOTE 2
NOTE 1:
RECOMMENDED METHOD
ALTERNATE METHOD
10
INCLUDES BLOWER DOOR TEST
LEGEND
3
A/C SYSTEM
?
Y
N
NOTE 2:
5
NOT RECOMMENTDED
NOTE 1
Figure D.1. Suggested flowchart for general residential commissioning.
63
Table D.1. Recommended residential commissioning procedures.
Task Description Protocol Equipment Est.
Time
Energy
Savings
Potential
1
Insulation
Inspection
Check that the insulation
installation in walls, attic and
floor is in accordance with
specifications.
ConSol 1999, "CIEE Final
Project Report: Protocols for
Energy Efficient Residential
Building Envelopes"
toolbox 30
min.
Medium
2 Window
Inspection
Check that the window type is
in accordance with
specifications
None in existence Handheld
spectrometer
(prototype
currently
available)
15
min.
Medium
3 Blower Door
Test and Leak
Detection
Determine the air tightness of
the building shell and
determine the location of those
leaks.
ASTM E779-99 and ASTM
E1186-87
Blower Door or
equivalent and
Smoke stick
15 -
45
min.
Medium
4 Moisture Test Check that construction details
will not lead to moisture
problems later on. The
inspector must be
knowledgeable about common
moisture problems in the
region.
No general standards exist.
Information Resource: Lstiburek
(1994) "Moisture Control
Handbook: Principles and
Practices for Residential and
Small Commercial Buildings"
Surface scanning
dielectric meter
45
min.
Health and
Safety
5 Delta Q Test Determine the building shell
leakage and duct leakage (at
operating conditions) using a
combined test.
Home Energy Magazine,
Sept/Oct. issue, pg.37
Blower Door or
equivalent
30
min.
High
6 Duct
Pressurization
Determine the duct leakage at a
given pressure across the ducts.
ASTM E1554-94 Duct Blaster or
equivalent
45
min.
High
7 AHU Airflow:
Fan-assisted
Determine the airflow across
the air handler fan b
y
ductin
g
CEE (2000) "Specification of
Ener
gy
-Efficient Installation and
Duct blaster or
equivalent
30
min.
Medium
64
Flowmeter all of the flow through a
calibrated fan.
Maintenance Practices for
Residential HVAC Systems"
Section 3.13.1
8 AHU Airflow:
Plate and Grid
Determine the airflow across
the air handler fan by inserting
a calibrated flow plate into the
filter slot.
Manufacturer's instructions, Calibrated Flow
plate
15
min.
Medium
9 AHU Airflow:
Sum-of-
Registers
Determine the airflow across
the air handler fan by adding
up the duct leakage plus the
flow out of the registers. This
can be done on either the
supply or return side of the
system.
CEE (2000) "Specification of
Energy-Efficient Installation and
Maintenance Practices for
Residential HVAC Systems"
Section 3.13.3
Fan assisted flow
hood
45
min.
Medium
10 AHU Airflow:
Temperature
Split
Determine the airflow across
the air handler fan by
measuring the temperatures on
either side of the heating
element, and the energy into
the element.
CEE (2000) "Specification of
Energy-Efficient Installation and
Maintenance Practices for
Residential HVAC Systems"
Section 3.13.2
Temperature
sensors
30
min
Medium
11 Register
Airflows
Determine the airflow into
each room in order to
determine that the heat/ cooling
load of that room will be met
by the HVAC system
ACCA Manual J, load
calculations to determine heating
and cooling load for each room
Fan assisted flow
hood
45
min.
Medium
12 Superheat Test Determine the correct
refrigerant charge for the
HVAC system
CEE (2000) "Specification of
Energy-Efficient Installation and
Maintenance Practices for
Residential HVAC Systems"
Section 3.14.1
Refrigerant gauge
set, temperature
sensors
1
hour
High
13 Subcool Test Determine the correct
refrigerant charge for the
HVAC system
CEE (2000) "Specification of
Energy-Efficient Installation and
Maintenance Practices fo
r
Refrigerant gauge
set, temperature
sensors
1
hour
High
65
Residential HVAC Systems"
Section 3.14.2
14 Gravimetric
Test
Determine the correct
refrigerant charge for the
HVAC system
CEE (2000) "Specification of
Energy-Efficient Installation and
Maintenance Practices for
Residential HVAC Systems"
Section 3.14.4