PERFORMANCE MONITORING OF ADVANCED TECHNOLOGY
WOOD STOVES: FIELD TESTING FOR FUEL SAVINGS,
CREOSOTE BUILD-UP AND EMISSIONS
VOL. I
CONEG^ Poiicy
Research Center
United States Environmental Protection Agency
New York State Energy Research and Development Authority
-------
The New York State Energy Research and Development Authority
(NYSERDA) is a public benefit corporation chartered by the New York
State Legislature. It is governed by a 13-member Board of Directors ap-
pointed by the Governor with the consent of the Senate. State Energy
Commissioner William D. Cotter is Chairman of the Board and the Chief
Executive Officer. A President manages the Authority's RD&D programs,
staff, and facilities.
As expressed in its enabling legislation, the underlying rationale for
establishing the Authority is:
... that accelerated development and use within the State of new
energy technologies to supplement energy derived from existing sources
will promote the State's economic growth, protect its environmental
values and be in the best interests of the health and welfare of the
State's population ...
The legislation further outlines the Authority's mission as:
... the development and utilization of safe, dependable, renewable
and economic energy sources and the conservation of energy and
energy resources.
The Authority's RD&D policy and program stress well-designed
research, development and demonstration projects, based on technol-
ogies with potential for near-term commercialization and application in
New York State. The Authority seeks to accelerate the introduction of
alternative energy sources and energy-efficient technologies and to im-
prove environmental acceptability of existing fuels and energy pro-
cesses. The Authority also seeks to ensure that Federal research pro-
grams reflect the needs of the State.
The use of New York contractors and an awareness of energy-related
growth opportunities are part of the Authority's effort to support in-
dustry in New York. Concentrating on these objectives ensures that
NYSERDA's RD&D programs will produce maximum benefits for the
citizens and businesses of New York, while attracting the participation
of both the private sector and the Federal Government.
NYSERDA derives its research and development revenues from an
assessment upon the intrastate sales of the State's investor-owned gas
and electric utilities. The Authority also derives income from the invest-
ment of retained earnings and leased property, as well as from bond
financings of pollution control facilities and special energy projects.
Further information about NYSERDA's RD&D programs may be
obtained by writing or calling the Department of Communications, New
York State Energy Research and Development Authority, Two
Rockefeller Plaza, Albany, N.Y. 12223; (518) 465-6251.
Mario M.Cuomo William D.Cotter
Governor Chairman
State of New York New York State
Energy Research and
Development Authority
-------
PERFORMANCE MONITORING OF ADVANCED TECHNOLOGY WOOD STOVES:
FIELD TESTING FOR FUEL SAVINGS,
CREOSOTE BUILDUP AND EMISSIONS
Vol. I
Final Report
Prepared for
NEW YORK STATE
ENERGY RESEARCH AND DEVELOPMENT AUTHORITY
Project Manager
Dr. Lawrence R. Hudson
and
CONEG POLICY RESEARCH CENTER, INC.
Project Manager
Steven J. Morgan
Technical Development Corporation
and
U.S. ENVIRONMENTAL PROTECTION AGENCY
Project Manager
Robert C. McCrillis
Prepared by
OMNI ENVIRONMENTAL SERVICES, INC/
10950 SW 5th Street, Suite 160
Beaverton, OR 97005
834-EIM-CE-86
Energy Authority
Report 87-26 November 1987
-------
NOTICE
This report was prepared by OMNI Environmental
Services, Inc. in the course of performing work
contracted for and sponsored by the New York
State Energy Research and Development Authori-
ty, the CONEG Policy Reseach Center, Inc. and
the U.S. Environmental Protection Agency (here-
after the "Sponsors"). The opinions expressed
in this report do not necessarily reflect those
of the Sponsors or the State of New York and
reference to any specific product, service,
process or method does not necessarily consti-
tute an implied or expressed recommendation or
endorsement of same. Further, the Sponsors and
the State of New York make no warranties or
representations, expressed or implied, as to
the fitness for particular purpose, merchant-
ability of any product, apparatus or service or
the usefulness, completeness or accuracy of any
processes, methods or other information con-
tained, described, disclosed or referred to in
this report. The Sponsors and the State of New
York and the contractor make no representation
that the use of any product, apparatus, pro-
cess, method or other information will not
infringe privately owned rights and will assume
no liability for any loss, injury, or damage
resulting from, or occurring in connection
with, the use of information contained, de-
scribed, disclosed, or referred to in this
report.
-------
ABSTRACT
This report presents the results of a two-year study in Vermont and New York
monitoring woodstove performance. The objective of the study was to determine the
effectiveness of catalytic and non-catalytic low-emission woodstove technology in
reducing wood use, creosote and particulate emissions. Measurements of wood use
and creosote accumulation in chimney systems were made in a total of 68 homes over
a period of two heating seasons. Forty-two of these homes were equipped with
instrumentation to measure particulate emissions and directly-measured wood use.
Catalytic woodstoves, catalytic add-on/retrofit devices and non-catalytic low-
emission stoves were provided by various woodstove manufacturers for use by
volunteer homeowners during the study period. Conventional technology stoves were
also included to provide baseline data.
Averaged results indicate that the low-emission non-catalytic stoves and catalytic
stoves had lower creosote accumulation, wood use, and particulate emissions than
the conventional technology stoves, although the range of values was quite large.
The reductions in particulate emissions by the catalytic and low-emission stoves
were not as great as could be expected based on laboratory testing. The large
number of variables affecting stove performance in "real world" conditions make
identifying causative factors difficult. Additional analysis of data and further
testing are currently planned.
-------
ACKNOWLEDGEMENTS
The contractor wishes to express thanks for the advice and guidance provided to this
research effort by the following individuals:
Skip Hayden, Canadian Combustion Research Laboratory
Brad Hollomon, New York State Energy Research and Development Authority
Larry Hudson, New York State Energy Research and Development Authority
Bob McCrillis, U.S. Environmental Protection Agency
Steve Morgan, Technical Development Corporation
Rich Poirot, Vermont Agency of Environmental Conservation
James Ralston, New York State Department of Environmental Conservation
David Reinbolt, Coalition of Northeastern Governors
The participation and generosity of in-kind contributors is also appreciated. These
include:
Members of the volunteer households
New York State Department of Environmental Conservation
Mew York State Energy Research and Development Authority
Vermont Agency of Environmental Conservation
Vermont Department of Health
Woodstove manufacturers who contributed their products to the study
-------
CONTENTS
Section
VOLUME I
1 BACKGROUND AND STUDY DESIGN 1-1
BACKGROUND 1-1
STUDY DESIGN 1-2
2 METHODOLOGY 2-1
CREOSOTE 2-1
WOOD USE 2-2
Woodpile Measurements 2-2
Scale Weighings 2-3
Home Owner Estimates 2-3
PARTICULATE EMISSIONS 2-4
Equipment 2-4
Probe Placement 2-6
Sampling Regime 2-8
Laboratory Procedures 2-8
Data Processing and Quality Assurance Procedures 2-10
Reported Values and Calculations 2-11
COMBUSTOR LONGEVITY INSPECTIONS 2-13
Inspection of Catalytic Combustors 2-13
Laboratory Testing of Field Combustors 2-14
-------
CONTENTS (Continued)
Section Page
3 RESULTS AND DISCUSSION 3-1
CREOSOTE 3-1
Stove Technology 3-1
Chimney System 3-9
Individual Installations 3-12
Stove Switching 3-19
WOOD USE 3-23
Stove Technology (Scale Weighings) ... . 3-23
Stove Technology (Woodpile Measurements) 3-29
Method Comparisons 3-29
PARTICULATE EMISSIONS, BURN RATE, AND FUELING DATA . 3-43
Introduction 3-43
Catalyst Operational Time 3-68
Fuel Load Data . 3-91
Particulate Emissions 3-96
CATALYST EFFECTIVENESS 3-100
Introduction 3-100
Combustor Replacement ..... 3-100
CATALYST LONGEVITY 3-104
Homes Using Existing Catalytic Stoves 3-104
Laboratory Testing of Field Combustors 3-105
Inspections 3-111
Combustor Replacements 3-117
Operator Factors 3-118
Stove Design 3-119
Combustor Factors 3-120
POM and TCO Emissions 3-121
-------
CONTENTS (Continued)
Section
ANALYSIS 4-1
INTRODUCTION 4-1
BURN RATE EFFECTS ON PARTICULATE EMISSIONS 4-2
Analysis of Data 4-2
Discussion by Stove Model 4-14
Catalytic Stoves 4-15
Add-on/Retrofits 4-18
Low-emission Stoves 4-21
FUELING EFFECTS 4-24
Fuel Loading Frequency Effects on Particulate Emissions . . . 4-24
Fuel Loading Frequency Effects on Burn Rate 4-31
Fuel Loading Frequency Effects on Average Fuel Load 4-38
CATALYST OPERATION TIME 4-44
Catalyst Operation Time Effects on Particulate Emissions . . 4-44
Catalyst Operation Time Effects on Burn Rate 4-48
Catalyst Operation Time Effects on Creosote Accumulation . . 4-53
ALTERNATE HEATING SYSTEM EFFECTS 4-58
Alternate Heating System Effects on Particulate Emissions . . 4-58
Alternate Heating System Effects on Burn Rate 4-65
CHIMNEY SYSTEM EFFECTS 4-71
Chimney System Effects on Creosote Accumulation 4-74
Chimney System Effects on Particulate Emissions 4-76
Chimney System Effects on Burn Rate 4-78
FIREBOX SIZE EFFECTS 4-80
ADVANCED TECHNOLOGY STOVE ANALYSIS 4-87
Catalytic Stoves 4-87
Add-on/Retrofits 4-96
Low-emission Stoves , 4-102
CONVENTIONAL STOVES ANALYSIS 4-107
Performance Discussion 4-107
vn
-------
CONTENTS (Continued)
Section
5 DISCUSSION AND CONCLUSIONS 5-1
GENERAL 5-1
WOOD USE AND CREOSOTE ACCUMULATION 5-1
PARTICULATE EMISSIONS .... 5-2
Stove Technology Groups 5-2
Stove Models 5-3
6 RECOMMENDATIONS 6-1
DATA REDUCTION/EXISTING DATA BASE 6-1
Detailed Graphics 6-1
Review of Field Studies 6-1
Evaluation of Stove Design Factors .... 6-2
ADDITIONAL FIELD STUDY .... 6-2
Stove Inspections 6-2
Additional Stove Testing 6-2
7 REFERENCES 7-1
APPENDIX A STUDY HOME CHARACTERISTICS A-l
VOLUME II-TECHNICAL APPENDIX (COMPANION DOCUMENT)
APPENDIX B CALCULATION PROCEDURES B-l
APPENDIX C QUALITY ASSURANCE C-l
APPENDIX D - GRAPHS OF STOVE TEMPERATURE, FLUE OXYGEN, FUELING
PRACTICES, AND HEATING SYSTEM USE D-l
VI 1 1
-------
ILLUSTRATIONS
Figure Page
2-1 AWES/Data LOG'r System 2-5
3-1 Creosote Accumulation by Stove Technology 3-8
3-2 Creosote Accumulation by Chimney Configuration 3-11
3-3 Comparative Creosote Accumulation: Group II Homes 3-22
3-4 Wood Use by Stove Technology (Scale Weighing Measurements) 3-28
3-5 Comparative Wood Use: Group II Homes (Woodpile Measurements) .... 3-46
3-6A Particulate Emissions (g/hr): Individual Sampling Periods—
Catalytic Stoves 3-69
3-6B Particulate Emissions (g/hr): Individual Sampling Periods—
Add-on/Retrofits 3-71
3-6C Particulate Emissions (g/hr): Individual Sampling Periods—
Low-emission Stoves 3-72
3-6D Particulate Emissions (g/hr): Individual Sampling Periods—
Conventional Stoves 3-73
3-7A Burn Rate (kg/hr): Individual Sampling Periods—Catalytic Stoves . . 3-74
3-7B Burn Rate (kg/hr): Individual Sampling Periods—Add-on/Retrofits . . 3-76
3-7C Burn Rate (kg/hr): Individual Sampling Periods—Low-emission Stoves . 3-77
3-7D Burn Rate (kg/hr): Individual Sampling Periods—Conventional Stoves . 3-78
3-8 Particulate Emissions (g/hr) by Stove Model 3-88
3-9 Particulate Emissions (g/hr) by Stove Technology 3-89
3-10 Particulate Emissions (g/kg) by Stove Technology 3-90
3-11 Performance Comparison by Stove Technology 3-99
IX
-------
ILLUSTRATIONS (Continued)
3-12A Catalyst Longevity—Home N32, Stove P, Combustor A 3-107
3-12B Catalyst Longevity—Home N03, Stove C, Combustor B 3-108
3-12C Catalyst Longevity—Home V07, Stove C, Combustor B ......... 3-109
4-1A Particulate Emissions (g/hr) vs. Burn Rate—Catalytic Stoves .... 4-3
4-1B Particulate Emissions (g/hr) vs. Burn Rate—Add-on/Retrofits .... 4-4
4-1C Particulate Emissions (g/hr) vs. Burn Rate—Low-emission Stoves . . . 4-5
4-1D Particulate Emissions (g/hr) vs. Burn Rate—Conventional Stoves . . . 4-6
4-2A Particulate Emissions (g/kg) vs. Burn Rate—Catalytic Stoves .... 4-7
4-2B Particulate Emissions (g/kg) vs. Burn Rate—Add-on/Retrofits .... 4-8
4-2C Particulate Emissions (g/kg) vs. Burn Rate—Low-emission Stoves . . . 4-9
4-2D Particulate Emissions (g/kg) vs. Burn Rate—Conventional Stoves . . . 4-10
4-3A Particulate Emissions (g/hr) vs. Fuel Loading Frequency—
Catalytic Stoves 4-25
4-3B Particulate Emissions (g/hr) vs. Fuel Loading Frequency—
Add-on/Retrofits 4-26
4-3C Particulate Emissions (g/hr) vs. Fuel Loading Frequency—
Low-emission Stoves 4-27
4-3D Particulate Emissions (g/hr) vs. Fuel Loading Frequency—
Conventional Stoves 4-28
4-4A Burn Rate vs. Fuel Loading Frequency—Catalytic Stoves 4-33
4-4B Burn Rate vs. Fuel Loading Frequency—Add-on/Retrofits 4-34
4-4C Burn Rate vs. Fuel Loading Frequency—Low-emission Stoves 4-35
4-4D Burn Rate vs. Fuel Loading Frequency—Conventional Stoves 4-36
4-5A Fuel Loading Frequency vs. Average Fuel Load—Catalytic Stoves . . . 4-39
4-5B Fuel Loading Frequency vs. Average Fuel Load—Add-on/Retrofits . . . 4-40
4-5C Fuel Loading Frequency vs. Average Fuel Load—Low-emission Stoves . . 4-41
4-5D Fuel Loading Frequency vs. Average Fuel Load—Conventional Stoves . . 4-42
4-6A Particulate Emissions (g/hr) vs. Catalyst Operation—Catalytic Stoves 4-46
4-6B Particulate Emissions (g/hr) vs. Catalyst Operation—Add-on/Retrofits 4-47
4-7A Burn Rate (kg/hr) vs. Catalyst Operation—Catalytic Stoves 4-50
4-7B Burn Rate (kg/hr) vs. Catalyst Operation—Add-on/Retrofits 4-51
4-8A Creosote Accumulation vs. Catalyst Operation—Catalytic Stoves . . . 4-54
4-8B Creosote Accumulation vs. Catalyst Operation—Add-on/Retrofits . . . 4.55
-------
ILLUSTRATIONS (Continued)
4-9A Particulate Emissions (g/hr) vs. Heating System Use—
Catalytic Stoves 4-60
4-9B Particulate Emissions (g/hr) vs. Heating System Use—
Add-on/Retrofits 4-61
4-9C Particulate Emissions (g/hr) vs. Heating System Use—
Low-emission Stoves 4-62
4-9D Particulate Emissions (g/hr) vs. Heating System Use—
Conventional Stoves . 4-63
4-10A Burn Rate (kg/hr) vs. Heating System Use—Catalytic Stoves 4-66
4-1OB Burn Rate (kg/hr) vs. Heating System Use—Add-on/Retrofits 4-67
4-10C Burn Rate (kg/hr) vs. Heating System Use—Low-emission Stoves .... 4-68
4-10D Burn Rate (kg/hr) vs. Heating System Use—Conventional Stoves .... 4-69
4-11A Particulate Emissions (g/hr) vs. Firebox Size—Catalytic Stoves . . . 4-81
4-11B Particulate Emissions (g/hr) vs. Firebox Size—Add-on/Retrofits . . . 4-83
4-11C Particulate Emissions (g/hr) vs. Firebox Size—Low-emission Stoves . 4-84
4-11D1 Particulate Emissions (g/hr) vs. Firebox Size—Conventional Stoves . 4-85
4-11D2 Particulate Emissions (g/hr) vs. Firebox Size—Conventional Stoves . 4-86
4-11E Particulate Emissions (g/hr) vs. Firebox Size—All Stoves 4-88
-------
TABLES
Table Page
1-1 Study Stove Categories 1-2
1-2 Study Stove Populations .... 1-5
2-1 Particulate Sampling Locations 2-7
3-1 Creosote Accumulation By Woodstove Technology Type .... 3-2
3-2 Creosote Accumulation By Chimney Configuration . 3-10
3-3 Creosote Accumulation By Stove Model 3-13
3-4 Effects Of Stove Technology Changes On Creosote Accumulation 3-20
3-5 Wood Use—Scale Weighing Measurements 3-24
3-6 Wood Use--Woodpile Measurements 3-30
3-7 Wood Use—Scale Weighing And Woodpile Measurements By Technology Type 3-36
3-8A Wood Use By Stove Model -- Catalytic Stoves 3-38
3-8B Wood Use By Stove Model -- Add-On/Retrofits 3-40
3-8C Wood Use By Stove Model -- Low-Emission Stoves 3-41
3-8D Wood Use By Stove Model -- Conventional Stoves 3-42
3-9 Effects Of Stove Technology Changes On Wood Use 3-44
3-10A Stove Use Characteristics 3-47
3-10B Fuel Characteristics 3.53
3-10C Emission Characteristics 3-59
3-11A Stove Use Characteristics By Stove Model 3-79
3-11B Emission And Burn Rate Characteristics By Stove Model 3-83
XI 1
-------
TABLES (Continued)
Table
3-12 Catalyst Operational Characteristics 3-92
3-13 "Student's t" Statistical Emission Rate Comparison 3-97
3-14 Effects Of Combustor Change On Particulate Emissions,
Burn Rate, And Catalyst Operation: Stove Code D 3-101
3-15 Laboratory Test Results: New Vs. Used Combustors 3-106
3-16 1985-1986 Heating Season Combustor Inspections 3-112
3-17 Combustor Replacement Chronology 3-118
3-18A POM And TCO Emissions (g/m3) 3-122
3-18B POM And TCO Emissions (g/hr) 3-123
3-19 POM And TCO Mass Fractions 3-126
4-1 Chimney System Effects On Creosote Accumulation,
Emission Rate, And Burn Rate 4-72
Xll 1
-------
SUMMARY
A study of woodstove performance was conducted during the 1985-86 and 1986-87
heating seasons in the Northeast. Sixty-eight homeowners in the Waterbury,
Vermont, and Glens Falls, New York, areas were provided with selected "advanced
technology" stoves or asked to use their existing (conventional) stoves for the
study period. The stoves were monitored for wood use, creosote accumulation in the
chimney system, and particulate emissions. Three advanced technology stove
categories (catalytic stoves, add-on/retrofit devices, and low-emission, non-
catalytic stoves) were compared with conventional technology stoves. Objectives of
the study were to evaluate the performance of the advanced technology stoves for
safety factors (creosote), efficiency (wood use), and environmental impacts
(particulate emissions). Special emphasis was placed on the effectiveness of
catalytic combustors.
Creosote and volumetric woodpile measurements were conducted on all 68 homes.
Creosote accumulation was measured by periodically sweeping the chimney system and
weighing the collected material. Wood use was monitored by measuring wood piles
during the heating season and normalizing for moisture content and fuel species.
Additionally, 34 homes were routinely sampled for particulate emissions over one-
week periods. These homes had data logging systems to record stove temperatures,
flue gas oxygen concentrations, and wood weights. Particulate samples consisted of
integrated samples collected every half hour during each week-long sampling period.
Flue gas flow rates were calculated based on combustion stoichiometry: burn rates,
fuel species, flue gas oxygen measurements, and estimated CO/C02 levels.
It is important to note that a large number of variables were found in field stove
installations: chimney systems, fuel characteristics, user practices, stove
maintenance, etc. The range of values recorded in all categories was quite large.
Reported data, while representing the values recorded during this study, may not be
representative of other climates, fuel woods, stove or catalytic combustor models,
chimney systems, or stove use patterns. Great care should be used in extrapolating
these findings to other circumstances.
S-l
-------
Due to the high variability and large range of data, averages from advanced
technology stove groups were, in most cases, not statistically different from the
conventional stove group. "Student's t" tests showed that only the low-emission
non-catalytic stove group had a mean particulate emission rate with a greater than
90% probability of being different and hence lower than those from the conventional
stove group. Emissions from individual stove models, however, were statistically
different from the mean of the conventional stoves in many cases. All advanced
technology devices (catalytic, add-on/retrofit, and low-emission non-catalytic)
showed lower average particulate emission rates, wood use, and creosote than the
conventional technology. Figure S-l summarizes averaged results from the stove
technology groups.
The stove technology group data represent averages, and reflect a wide range of
values. In general, all stove categories, including conventional stoves, had
models and specific installations with low (and high) particulate emissions. It is
therefore most appropriate to evaluate stove performance on a model-by-model basis,
recognizing that due to the relatively small number of installations and stove
models, values may not be representative of "typical" stove performance.
Even though the number of individual samples is high, the wide range of values and
the large number of variables makes identifying causative factors difficult.
Results presented in this report are from a number of different stove types and
models in different installations, in which homeowners used different fuels and
operating procedures. A thorough review of stove burn rates, fuel loading
practices, catalyst operation time, and frequency of alternate heating systems did
not identify a single factor responsible for emission patterns. This indicates
that while many factors can affect particulate emission rates, no single factor
appears to be dominant in all stove types or models. In general, however, it
appears that stoves with smaller fireboxes, regardless of technology type, tend to
have lower emission rates.
General conclusions are presented below in the following categories: Advanced
Technology, Catalyst Performance, Operator Practices, Technology Factors, and Other
Findings.
1.0 Advanced Technology Performance
1.1 Most stoves in the advanced technology categories (catalytic, add-
on/retrofit, low-emission non-catalytic) episodically demonstrated
S-2
-------
Figure S-l
Performance Comparison by Stove Technology
I
co
30.0 -
25.0 -
HEAN
PARTICULATE
EMISSIONS
(G/HR)
15.0 -
10,0 -
5.0 -
ADD-ON/
RETROFITS
CQNUENTIONflL
STDUES -r
CATALVTIC
STDUES
LOW EMISSION
STOUES
PRRTICULflTE EMISSIONS (G/HR)
WOOD USE (KG/HDD)
CREOSOTE flCCUHULAHDN (KG/1000 HDD)
+ 1 SD
~j HEAN
| -1 SD
- 2.00
- 1.10
- 1,50
MEAN
- UDDD USE
(KG/HDD)
- 1.20
MEAN
CREOSOTE
ACCUMULATION
(KG/1000 HDD)
- 0,50
- O.tO
- 0.20
-------
lower emissions than the baseline conventional stoves under "field use'
conditions. Good performance in at least one installation for most of
the stove models indicates that factors, such as stove maintenance and
fueling practices, may be as important as stove technology features in
achieving low emission rates. Stove firebox size, regardless of
technology group, was a prime factor in determining emission rates;
smaller stoves had lower emissions.
1.2 In general, performance of the stove technology groups appeared to
be consistently ranked in terms of particulate emission rates, wood
use, and creosote accumulation; low-emission non-catalytic stoves
had the lowest particulate emission rate, wood use, and creosote
accumulation, while conventional stoves had the highest. It should
be noted that only low-emission non-catalytic stoves showed a mean
emission rate which was statistically different from the
conventional stoves. It should also be noted that creosote
accumulation is strongly influenced by flue system type and wood use
appears to be influenced by burning patterns and firebox size.
1.3 All advanced technology stove groups averaged lower wood use and
creosote accumulation rates when households switched from
conventional stoves between heating seasons. Average reductions by
stove group ranged from about 10% to 35% for creosote and from about
15% to 30% for wood use.
1.4 The low-emission stoves, as a group, had the lowest average
emissions. Each model had different burning characteristics; most
showed relatively good performance. Average results from this
technology group are strongly influenced by the good performance of
two EPA 1990-certifiable stoves (M and N). Furthermore, excluding
one high-emission home (V18, using non-EPA-certified Stove K) would
reduce average emissions in this category from 13.4 to 10.0 g/hr,
and reduce the standard deviation (o-) from 10.2 to 5.7.
1.5 User satisfaction was generally high with the advanced technology
stoves provided to study homes. In particular, homeowners with
catalytic and low-emission stove models were frequently pleased with
the units. (In some cases, user satisfaction remained high even
though the catalytic combustor had deteriorated.) Some add-on
devices also received positive comments. The add-on with the lowest
average particulate emission rate also received homeowner complaints
about smoke spillage.
2.0 Catalyst Performance
2.1 Catalytic stoves showed variable performance. Most individual
models performed well in some homes. Other installations had
relatively high emissions. Overall, performance of these stoves did
not match the expectations created under ideal laboratory
conditions, although only one of the catalytic models was EPA 1990
certifiable. The mean emission rates of existing catalytic stoves
and new catalytic stoves were virtually identical. User education
and further technology refinements remain possible factors which
could help improve the performance of catalytic stoves.
2.2 Add-on/retrofit devices did not perform well overall, but 2 devices
reduced emissions considerably. The stoves on which these devices
S-4
-------
The high degree of variability in performance and the relatively
small sample populations make comparisons inappropriate.
5.2 Conventional stoves in this study may be cleaner-burning heaters
than are "typical." Four of the six conventional stoves had
relatively small fireboxes (< 2.4 ft^), and two of these had small
effective fireboxes (< 1.5 ft^). Emissions from these stoves
therefore may not be typical of existing stove technology.
Additionally, the cold Northeast climate and commensurately higher
burn rates preclude direct comparison to stove performance in milder
climates.
5.3 Alternate heating system use did not correlate well with particulate
emission rates or burn rates, although heating system use was
monitored only in the room with the stove. In general, most homes
in the study used their alternate heating system less than 3.5% of
the time (while the stove was operating). This amounts to less than
one hour per day. A large portion of the homes used no back-up heat
at all.
5.4 Polycyclic organic material (POM) emissions were variable and non-
conclusive. Testing method and analytical method limitations, and a
very limited database, preclude any ranking of POM emissions by
stove type.
S-7
-------
Section 1
BACKGROUND AND STUDY DESIGN
BACKGROUND
The use of catalytic combustors in reducing particulate emissions from woodstoves
has been shown to have considerable potential, based on laboratory test results.
In recognition that the combustors would likely experience some loss of
effectiveness over time and that "real world" conditions would have an unknown
effect on combustor performance, documentation of catalytic woodstove performance
was sought, A consortium of funding partners, comprised of the Coalition of
Northeastern Governors (CONEG), New York State Energy Research and Development
Authority (NYSERDA), and U.S. Environmental Protection Agency (EPA), sponsored a
two-heating-season study to investigate the effectiveness of "advanced technology"
woodstoves.
Direct project funding was provided by CONEG, NYSERDA, and EPA. In-kind
contributions of services were provided by New York State Department of
Environmental Conservation (NYSDEC), Vermont Agency of Environmental Conservation
(VAEC), and Vermont Department of Health (VDOH). Woodstoves were provided by
various stove manufacturers. Stoves were placed in the homes of volunteer
participants.
The study objictives were to evaluate the performance of several types of stove
technology under typical use conditions for:
• safety (creosote reduction)
• efficiency (wood use reduction)
• environmental impacts (particulate emission reduction)
It should be noted that the objectives were not to demonstrate the potential for
advanced technology woodstoves, but to document typical performance of available
(fall 1985) technology in the Northeast region.
1-1
-------
STUDY DESIGN
Six stove technologies (Table 1-1) were selected for investigation, representing
residential natural draft wood-burning devices.
Table 1-1
STUDY STOVE CATEGORIES
Stove
Technology
Type
Stove
Model
Types/
(Codes)
Comments
Catalytic
(A,B,C,D)
Four manufacturers provided new stoves for the study.
Catalyst stoves are defined as having the combustor as
an integral part of the new stove. Existing catalytic
stoves (Group III homes) represented five additional
models.
Add-on
Add-on devices are defined as units which can be added
to virtually any stove at the flue collar. Three
devices were used for the first year of the study, and
one was added for the second year.
Retrofit
2
(E,F;
Retrofit devices are designed to fit one stove model or
design type, and typically are close-coupled to the
stove.
Low-
Emission
Non-
Catalytic
(K,L,M,N)
"Low-emission" stoves are defined for this study as
non-catalytic models which have been certified under
the Oregon DEQ program. Two stove models were included
for the first year of the study, and two more
"EPA-1990-certifiable" models were added for the second
year.
'onventiona 1
6
(0)
These are defined as existing stoves in study homes,
representing a range of designs. They are generally
categorized as typical of conventional woodstove
technology.
ixisting
Catalytic
6
(P)
These are defined as existing catalytic stoves in study
homes with one to two heating seasons of prior use.
One stove was the same model as one used in the
catalytic group.
1-2
-------
New woodstoves were provided to the study by woodstove manufacturers.
Contributions of stoves and shipping/installation costs were solicited from
producers making stoves which had passed or were capable of passing the Oregon
Department of Environmental Quality (DEQ) Woodstove Certification Program. Stoves
were provided in the fall of 1985 by manufacturers interested in participating in
the study. Of the catalytic stoves, one met DEQ 1988 standards ("weather-weighted"
average emission rate of 4.0 grams per hour), one was subsequently certified to
1986 standards ("weather-weighted" average emission rate of 6.0 grams per hour),
and one, while never certified, appeared capable of meeting the 1988 standard,
based on limited certification-type testing. The fourth catalytic model, while
never tested in a laboratory, was a prototype of a model certified to DEQ 1986
standards. (The secondary air system was modified for the production model.)
Three catalytic add-on devices were originally used in the study. Add-on devices
are not covered by current Oregon DEQ or U.S. EPA woodstove regulations, but
research testing had been conducted on two of the three units. At the beginning of
the second heating season, a fourth add-on device was added to the study, based on
lab tests showing this unit to have the best emission reduction potential of tested
add-on devices.
The two catalytic retrofit devices had both been certified to Oregon 1986 catalytic
standards. One of the retrofit models was discontinued subsequent to its inclusion
in the study. For purposes of analysis, add-ons and retrofits were considered as
similar technologies; they would both be available for installation on existing
woodstove installations and thus have the potential to reduce emissions from
existing stoves.
All of the catalytic stoves and add-on/retrofit devices were equipped with
combustors supplied by the stove manufacturer. Combustors were manufactured by
Applied Ceramics, Corning, or Panasonic (Technical Glass Products). The three
combustor makes were approximately equally represented in the catalytic devices.
Two low-emission stove models were included in the first heating season. One of
these stoves met DEQ 1986 non-catalytic stove standards ("weather-weighted" average
emission rate of 15.0 grams per hour), and one met DEQ 1988 standards ("weather-
weighted" average emission rate of 9.0 grams per hour). Based on preliminary
indications that this technology group may perform relatively well in the field,
two more models were added at the beginning of the 1986-87 heating season. Both of
1-3
-------
these new models were certified to DEQ 1988 standards and should be capable of
meeting U.S. EPA 1990 non-catalytic standards ("national weather-weighted" average
emission rate of 7.5 grams per hour).
No special training was provided to homeowners regarding proper operation of the
advanced technology stoves. Stoves were installed by professional installers, who
answered questions but did not attempt to train homeowners. An instruction manual
for each stove was left with the homeowners.
All of the stove models are coded in this report to provide anonymity to
manufacturers who provided or donated equipment to the study. This is in
recognition of their accepting the risk that, for whatever reasons, their product
may not have performed as expected.
Stoves were installed in volunteer households selected from a list of applicants
provided by VAEC and NYSDEC. Potential participants were interviewed, and the
homes and existing woodstove systems inspected. Homes were evaluated for occupant
enthusiasm for the project, chimney size and venting characteristics (to match with
available stoves), geographic location, and other factors. A total of 66 homes
were initially selected for the study; 33 in Glens Falls, New York, and 33 in the
Waterbury, Vermont, area. All homes used wood as a primary heat source.
Manufacturers offered homeowners a discount on buying the stove at the end of the
study, or gave the appliance to the homeowners. All participants received chimney
sweeping services free of charge during the study. Two homes were added to the
study group for the 1986-87 heating season due to original participants dropping
out of the study.
The study homes were divided into three groups, each receiving varying levels of
investigation. Group I homes, totaling 32 with 16 in each state, were monitored
for creosote accumulation, woodpile use, and particulate emissions. With the
exception of isolated participant dropouts, most Group I homes continued to use the
same stove through both heating seasons. Some Group I homes changed to low-
emission stove models for the second heating season as part of an emphasis shift in
the study. Each Group I home was scheduled for seven emission sampling periods.
An additional four homes in this group were monitored for creosote accumulation and
wood use while serving as backup homes in case a Group I home dropped out of the
study.
1-4
-------
Group II homes, totaling 24 with 12 in each state, received creosote accumulation
and woodpile use measurements. No particulate data were collected. Group II homes
switched stoves between heating seasons to allow comparisons of creosote-
accumulation and wood use, with the house, chimney, and occupants remaining
constant.
Group III homes, totaling six with three in each state, were monitored for creosote
accumulation, woodpile measurements, and particulate emissions. Group III homes
already had catalytic stoves which had been in use for at least one heating season
prior to this study. Emissions were measured once during the first heating season
on all six homes and once during the second heating season on four homes.
Log books were left in all homes for occupants to record unusual events or
occurrences.
Table 1-2 lists the stove technologies in each study group for the two heating
seasons.
Table 1-2
STUDY STOVE POPULATIONS
Group I
Group II
Group III
Total
Catalytic
'85-86 '86-87
14
3
6
23
14
6
6
26
Add-on/Retrofit
'85-86 '86-87
12
0
0
12
6
7
0
13
Low-Emission
'85-86 '86-87
3
0
0
3
10
3
0
13
Conventional
'85-86 '86-87
7
21
0
28
5
5
0
10
The shift in the types of Group I stove technology between heating seasons was due
to reducing the number of add-ons and increasing the number of low-emission stoves.
Based on relatively high emissions from most add-on devices and the discontinuation
of one of the retrofit devices (F) by the manufacturer, many of these devices were
1-5
-------
pulled from the study. They were replaced by an add-on device which had performed
well in laboratory tests (J) and two models of low-emission stoves (M,N) which were
considered to be among the best available non-catalytic stove designs.
Group II homes were scheduled to run one heating season with one stove and the next
with another stove, as described previously. This was conducted as planned. Group
III homes were also tested as planned.
1-6
-------
Section 2
METHODOLOGY
CREOSOTE
Documentation of creosote accumulation in study home flues was conducted by
measuring the net amount of material removed from each chimney by "sweeping" with
brushes. Chimneys were swept at the beginning, middle, and end of the 1985-86
study by professional chimney sweeps. More frequent sweepings were conducted on an
as-needed basis during the 1986-87 heating season due to concern over the potential
for chimney fires.
Creosote dislodged by the sweeping was collected by the chimney sweeps. The first
sweeping in Fall 1985 was to establish "clean" conditions, while creosote collected
in subsequent sweepings was collected and weighed. Weighing was conducted by OMNI
field personnel. The mass of creosote collected was then normalized by heating
degree-days (HDD) occurring during the creosote accumulation period between
sweepings. (Creosote samples from the first mid-season [1985-86] sweeping of all
study home chimneys were sent to the Solar Energy Research Institute [SERI] for
chemical analysis. The analysis was conducted independently of this project, and
therefore no results of chemical composition are reported here.)
The heating "load," in HDD, during the study period was calculated for each home.
Heating degree-days were summed for the period between creosote sweepings, yielding
the heating load for the specific creosote accumulation period for each home. Only
heating degree-day data from October 15 through April 30 of each heating season
were used in the heating load calculations, as little woodstove use was reported
outside this time period.. Weather data were from the Waterbury and Glens Falls
weather recording stations maintained by the Northeast Regional Climate Center.
Glens Falls data were used for New York homes and Waterbury data were used for
Vermont homes.
The same chimney sweeping firm conducted the chimney cleanings during both heating
seasons in Vermont. A total of four sweeps from this firm were involved in
cleaning Vermont study home chimneys. A single sweep conducted all sweepings in
New York during the first heating season, while two additional sweeps were used
during the second heating season.
2-1
-------
It should be noted that this methodology was intended to provide qualitative
indications of creosote deposition. Chimney type and configuration,
revolati1ization of deposited material, liquid condensation, and other factors
preclude using this data for purposes other than as general indicators of creosote
accumulation.
WOOD USE
Woodpile Measurements
Wood use was measured at all study homes by monitoring the dimensions of
participants' woodpiles. Total overall woodpile dimensions were recorded at the
beginning, middle, and end of each heating season. Any additions or non-burning
removals were noted. Wood moisture, species, and species mix were documented
during each visit to the study homes. Wood moisture was measured with a
resistance-type meter using insulated pins. Fuel species and species mix
determinations were made by field personnel. Instrumented (Group I) homes were
visited twice per month, while Group II and III homes received three measurements
during the course of the year.
At the completion of each heating season, the net volume of wood used during the
winter was calculated. The mix of wood species was used to calculate a species-
weighted wood density, based on standard published values. A stacking density (the
ratio of actual wood volume to woodpile volume) of 0.66 was assumed, based on field
measurements and published data (_!_). All "cord densities" were then normalized to
zero percent moisture. The dry cord density for each study home woodpile was
multiplied by the volume of wood consumed during the measurement period, yielding
the total mass of dry wood burned.
The heating load during the study period, measured in heating degree-days (HDD),
was calculated for the Vermont and New York areas. Heating degree-days were summed
for the period between woodpile measurements, yielding the heating load for the
study period specific to each home. Heating degree-day determinations were made
using the same criteria used in normalizing creosote accumulation data.
Woodpile measurements, while providing an overall indication of total wood
consumption during the year, do not address how often the stove is used, average
burn rate, home weatherization, or how often an alternative heat source was used.
2-2
-------
Scale Weighings
In Group I homes, wood use was measured with an electronic scale attached to a
woodbasket. These measurements were made in conjunction with particulate emission
measurements. Homeowners were instructed to place wood in the scale basket prior
to loading the stove. Before placing fuel in the firebox, the homeowner would
depress one of four buttons corresponding to existing coalbed conditions. Fuel
would then be placed in the stove, and an "ENTER" button pushed, recording the net
weight of wood placed in the stove. Data were recorded by a data logging system
described below. Scales were calibrated at the beginning and end of each recording
period. A more detailed description of this equipment is provided on the following
page.
Group I and Group III homes recorded wood use using the scale/data logger system
during particulate sampling periods. In an effort to maintain the diligence of
participants and prevent "burn-out" in using the system, homeowners were asked to
use woodbaskets only during sampling periods. Moisture and species of wood in the
woodbasket were measured and recorded by OMNI field staff twice per week during
sampling periods. The mass of wood burned during the week-long sampling period was
corrected to zero percent moisture and normalized using the total HDD during the
recording period, as described above.
Home Owner Estimates
At the end of the 1985-86 heating season, homeowners were asked to estimate their
wood use over the past winter. A written survey asked homeowners, among other
questions, "How many cords of wood did you burn last winter?". Completed surveys
were mailed to OMNI and results were tabulated. However, woodpile measurements
were made in December 1985 after the wood heating season had been under way for at
least a month. Accurate comparisons were therefore not possible, although it
appeared that, in general, homeowners overestimated their wood use when compared to
measured wood use.
An informal verbal survey by OMNI field staff at the end of the 1986-87 heating
season in Vermont showed that on a qualitative basis, homeowners again tended to
overestimate wood use, although considerable variation was noted.
2-3
-------
PARTICULATE EMISSIONS
Equipment
While extensive data are available on stove performance measured in a laboratory,
virtually no data have been available on woodstove emissions under actual "field"
conditions. Laboratory tests are conducted under a set of rigorously controlled
conditions which minimize the variables that can affect emission values (2.). Field
conditions necessarily include such variable factors as fuel species, moisture,
piece size, loading density, fueling frequency and burn rates, chimney system
configurations, and stove operation factors. With catalytic stoves, additional
factors such as bypass operation and catalyst "preheating" practices can be
significant.
Particulate emissions were measured with a pair of instruments developed by OMNI
for field measurements of woodstoves. Particulate samples were collected in an
Automated Woodstove Emission Sampler (AWES). Wood use, flue gas oxygen, and
various temperature values were recorded by a programmable microprocessor/
controller dubbed the Data LOG'r. A schematic of the system is shown in
Figure 2-1.
AWES Description. The AWES sampler was specifically designed for sampling
residential woodstove particulate emissions. As programmed in this study, it was
capable of sampling woodstove emissions for periods up to one week in length.
Sample flow was maintained by a critical flow orifice, so no adjustment was
required during operation. Sample start and stop times, dates, and frequency
(minutes on and minutes off) were programmable and controlled by the Data LOG'r.
Each sampler was installed prior to scheduled start time, left unattended, and
removed for sample processing at the end of the sampling period.
Each AWES unit drew flue gases through a stainless steel probe, Teflon tubing, and
a U.S. EPA Method-5-type filter (heated to about 75-115°C) for collection of
particulate matter, followed by an adsorbent resin (XAD-2) trap for semi-volatile
hydrocarbons. Water vapor was removed by a silica gel trap. Flue gas oxygen con-
centrations, which are used in conjunction with wood use data to determine flue gas
volumes, were measured by an electrochemical cell. The AWES units use a critical
orifice to maintain a nominal sampling rate of 1.0 liters per minute (0.035 cfm).
Each AWES critical orifice was calibrated to determine the exact sampling rate.
Appendix C shows data on AWES equivalency to other reference methods.
2-4
-------
ro
en
Figure 2-1
AWES/Data LOG'r System
SS3
OUT-
DOORS
SS2_
ROOM"
ssi
\ MEMORY
CARTRIDGE
Z PFOGRAMASLE
SOFTWARE
3. AUX BATTERY
PACK
4. FAILURE ALARM
Data LOG'r
WOOD SCALE
EXHAUST RETURN
THERMOCOUPLE 1
INLET
HEATED
CHAMBER
J
T
T
T
WOOD STOVE
AWES
-------
Data LOG'r Description. The Data LOG'r is a multi-channel programmable
microprocessor/controller with the capability of processing both digital and analog
signals. The unit has data storage capacity of 32 kilobytes on a field-
replaceable, non-volatile memory data cartridge. As programmed for this study,
cartridge capacity allowed up to 30 days of continuous operation between servicing
in most field project applications. The Data LOG'r was programmed to record and
store the following information:
• Starting date, time, and unit serial number for data recording
periods;
o Daily date and time, recorded at midnight, and a continuous
time record in five-minute intervals;
• Flue gas, in-catalyst, and before-catalyst temperatures (where
applicable) averaged over 15-minute intervals;
• Record of alternate heating system status (on or off) by use of a
temperature sensor;
• Wood weights and coalbed condition, recorded when the woodstoves
were fueled;
• Oxygen measurements when the AWES units were sampling, recorded
every 30 minutes; and
• Home VAC power status (on or off), measured at five-minute
intervals.
The attached electronic scale/woodbasket units supplied an analog voltage output
proportional to the weight placed in the wood holder. Scale readings were recorded
by having the homeowners use an attached keypad in a prescribed sequence. The
keypad also allowed the homeowners to record the coalbed conditions at each time of
stove fueling.
The Data LOG'r was programmed to activate the AWES unit(s) at a specific date and
time. Sampling intervals were one minute every 30 minutes for seven-day sampling
periods, commencing on Saturdays at midnight.
Probe Placement
AWES sampling probes were located at several points in the stove/flue system during
the first heating season. All stoves were sampled at the flue collar for
conventional, catalytic, and low-emission stoves, and at the exit of the add-on
devices. AWES probes were placed 0.3 meters downstream from the flue collar or
add-on unit. This permitted direct comparison of stove performance without chimney
2-6
-------
deposition effects. Modifications were also made to the probe placement plan for
the second heating season, as noted in Table 2-1, based on field experience and
results from the first heating season.
Table 2-1
PARTICULATE SAMPLING LOCATIONS
Heating
Season
1985-86
1987-87
Homes
Sampled
38
36
Firebox
8
4
Flue
Collar
38
36
After
Add-on
8
4
Chimney
Exit
12
0
Additional
Flue Collar
0
12
Firebox samplers were reduced in number due to secondary air introduced into the
stoves between the pair of AWES samplers, while only the flue collar AWES was
recording 02. The resulting dilution of the sample affected all reported
particulate values. Some firebox samplers were left in the study for the second
heating season to allow reporting of organic compounds as a fraction of the total
particulate catch. The total number of add-on devices in the study was reduced for
the second heating season.
Chimney samplers were eliminated after the first heating season due to problems
encountered with dilution of samples from leaking flue systems (only the flue
collar AWES had an oxygen sensor), freezing sample lines, and dangerous working
conditions on the rooftops.
Due to questions of the accuracy and representativeness of firebox and chimney exit
samples, results from these sampling locations are not presented in this report.
«
Additional samplers were added at the flue collar for the second heating season in
11 homes to document AWES sampler precision in the field.
2-7
-------
Sampling Regime
Emission sampling equipment was installed by OMNI personnel in study homes during
the week preceding a sampling period. Wood moisture content measurements and
species determination were recorded from wood placed by the stove. All wood
moisture measurements were performed using a Delmhorst moisture meter (Model RC-
1C) with insulated pins. The participants were given instructions on the operation
of the Data LOG'r keypad/scale unit and provided with a log book for recording
unusual events.
Field staff visited each study home to service the sampling equipment at the start
and end of each sampling period. At the start of a sampling period, the AWES unit
was installed; leak checks were performed; thermocouples, the woodbasket/scale
unit, and the oxygen cell were calibrated; the Data LOG'r was programmed with the
proper sampling interval and start/stop times; and wood moisture measurements were
performed on the fuel in the home's storage area. At the end of each sampling
period, end-of-sampling-period calibrations and leak checks were performed; the
AWES unit, sampling line, and sampling probe were removed; and wood moisture
content and species were recorded as before.
The Data LOG'rs were programmed to activate the AWES units for one minute per half
hour for seven days. Study homes were sampled every four weeks. Two groups were
located in Vermont and two in New York. Homes in the two states were sampled
sequentially. For example, during the week while Group A (Vermont) woodstove
emissions were being sampled, field personnel installed the AWES units and sampling
probes in the Group B (New York) homes. All sampling periods commenced on Sunday
at 0000 hours.
Laboratory Procedures
Each AWES unit was cleaned and prepared with a new filter and a purified XAD-2
adsorbent resin cartridge prior to field installation. After each sampling period,
the stainless steel sampling probe, Teflon sampling line, and AWES unit were
removed from the study home and transported to a laboratory for processing.
Laboratory facilities at the Vermont Agency of Environmental Conservation (1985-86)
and the Vermont Department of Health (1986-87) were used for AWES preparation and
sample recovery. Prior to transporting the AWES unit, the sample intake port,
sampling line, and sampling probe were sealed. The components of the AWES samplers
were processed as follows:
-------
1. Filters: Glass fiber filters were removed from the AWES filter
housings and placed in petri dishes. The petri dishes were sealed
and shipped to OMNI's Oregon laboratory for desiccation and
gravimetric analysis for particulate catch.
2. XAD-2 Adsorbent Resin: Resin cartridges were capped and shipped to
OMNI's Oregon laboratory. In the laboratory the cartridges were
extracted in a Soxhlet extractor with dichloromethane (SEMI grade)
for 24 hours. The extraction solvent was transferred to a tared
glass beaker. The solvent was evaporated at ambient conditions, the
beaker and residue desiccated, and the extractable residue weight
determined. The purified XAD-2 resin remained in the cartridge and
was reused.
3. AWES Hardware: All hardware exposed to the sample stream (stainless
steel probe, Teflon sampling line, glass filter housing, and all
Teflon, glass and stainless steel fittings) was rinsed with
dichloromethane (SEMI grade) and methanol (reagent grade). The
solvents were placed in 500 ml amber glass jars with Teflon-lined
lids which were capped, sealed, and shipped to OMNI's Oregon
laboratory. In the laboratory, the solvents were placed in tared
glass beakers. The solvents were evaporated at ambient conditions,
desiccated, and weighed to determine the residue weight.
After cleaning, the AWES units were reassembled for field use. The intake port,
sampling probe, and sampling line were sealed for transportation to the study home,
and unsealed immediately prior to installation.
POM/TCP Analysis. A subset of AWES samples was selected for analysis to determine
concentrations of specific polycyclic organic materials (POM) and total
chromatographable organics (TCO). Ten samples from the first heating season were
analyzed for POM concentrations. The samples submitted for testing consisted of
material extracted from the XAD-2 resin only. Although the specific POM compounds
were selected as indicators of the total POM concentrations, concerns were raised
regarding the possibility of significant POM concentrations in other portions of
the total AWES sample (solvent rinses or filter). POM samples submitted for
analysis during the second heating season were combined dichloromethane (CH2C12)
rinses, filter extracts (CH2C12), and XAD-2 extracts (CH2C12). POM compounds
selected for analysis were based on previous EPA research: naphthalene,
acenaphthene, phenanthrene, pyrene, benzo(a)pyrene, indeno(l,2,3-c,d)pyrene,
benzo(g,h,i)perylene and 3-methyl cholanthrene. POM analysis was conducted using a
gas chromatograph/mass spectrometer (3.).
2-9
-------
TCO analysis was conducted using a gas chromatograph with a flame ionization
detector (GC/FID) (4_). Hydrocarbon compounds with boiling points between 100°C and
300°C were reported.
It should be noted that the AWES sampler was designed for gravimetric measurements.
Several factors may influence the representativeness of reported POM/TCO values:
« Samples were at ambient temperatures (except the heated filter) in
the sampler. Material collected at the beginning of a sampling
period was therefore not recovered for 8-10 days.
• Samples were shipped from the field lab to OMNI's Oregon lab by air
freight at ambient temperatures.
• Analytical procedures used for identifying POM compounds are, under
the best conditions, relatively imprecise.
POM and TCO results were intended to provide a qualitative assessment of emissions
characteristics from the various stove technologies.
Data Processing and Quality Assurance Procedures
Using a portable computer, data files stored in the Data LOG'r memory cartridges
were downloaded in the field onto floppy disks at the conclusion of each sampling
period. The files were copied in the field office and one copy shipped to OMNI's
Oregon office. Each data file was reviewed to check for proper equipment
operation. Data LOG'r files were used in conjunction with the AWES particulate
sample and wood moisture data to calculate particulate emission rates, catalyst
lightoff times (when applicable), stove operation time, overall thermal efficiency,
and burn rates.
The Data LOG'r data files, log books, and records maintained by the field staff
were frequently reviewed to ensure sample integrity. Any parameter or calibration
objective that did not meet OMNI's in-house quality control criteria was rejected
or flagged and noted. The emission rate values that incorporated a flagged quality
assurance parameter were carefully reviewed and are footnoted in the data tables.
No flagged data were used in data summaries or comparisons of stoves or technology
groups.
2-10
-------
Participate emission rates were calculated with precision and accuracy values.
Each individual measurement that was used in the emission calculations has some
degree of uncertainty associated with it, and these uncertainties are propagated to
determine the precision and accuracy attached to each calculated particulate
emission rate. Appendix B lists the calculation procedures used for particulate
emission rate determinations. Appendix C summarizes the criteria used in the
precision and accuracy calculations.
Field blanks were collected with the AWES units to evaluate potential particulate
contamination of the AWES components, fittings, and sampling lines. The field
blank AWES units were prepared according to normal sampling protocols, leak
checked, left unattended for one week without being programmed to sample, leak
checked, and returned to the laboratory for sample processing. The mean
particulate catch from field blanks was subtracted from the total particulate catch
for each emission sample. Details of field blank factors are presented in
Appendix C.
Audits of Data Quality (ADQ), a Technical Systems Audit (TSA), and a Performance
Evaluation Audit (PEA) were conducted by an EPA-assigned auditor during the course
of the project. Audit results are presented in Appendix C.
Reported Values and Calculations
All the data reported, unless otherwise noted, represent samples obtained at the
flue collar (for catalytic, retrofitted, low-emission, and conventional stoves) or
above the add-on device. This allows direct comparison of the stove technology
groups without introducing direct chimney system effects. When duplicate samplers
were used, .this is noted, and an average of the two AWES samplers is reported,
based on the flue 0;? readings from one of the samplers. (Data LOG'rs used in this
study had only one 02 recording channel.)
Emission data are presented in the following formats:
• grams particulate/hour
• grams particulate/kilogram dry wood burned
• grams particulate/lO^ joule energy released into home
• grams particulate/m^
2-11
-------
Data presented in this report were calculated as a function of stove operation time
(stack temperature above 100°F at 0.3 meters above the flue collar or add-on
device). Values therefore represent emissions when the stove was in operation.
Emission data in the gram/kilogram format were calculated using the following
inputs:
1. Mass of participate material collected by the sampler.
2. Measured flow rate of the sampler (calibrated orifice, flowmeter).
3. Sampling duration (minutes of actual sampler pump operation).
4. Stoichiometric volume of gas produced by burning a known mass of
wood. This is a calculated value based on the elemental composition
of the wood fuel and flue gases. Specific values for carbon,
hydrogen, oxygen, and nitrogen were obtained from available
literature for 20 species of wood, and Stoichiometric gas volumes
were calculated based on the mix of fuel species burned, moisture
content, and burn rates. Average flue concentrations of CO and C02
were assumed based on technology type, and are shown in Appendix B.
5. Concentrations of oxygen in the flue (at the sampling location),
measured by an oxygen sensor cell in the sampler. Excess air was
calculated relative to ambient oxygen concentration.
When emissions were calculated in the gram/10^ J format, additional input data were
required:
6. Heat content of dry wood (J/kg). This was also obtained from
literature values for 20 species and calculated based on the mixture
of fuel used during the sampling period.
7. Stove efficiency. This was calculated for each sample based on
stack gas temperature, fuel moisture, excess air, and particulate
mass using the "Condar method." Details are presented in
Appendix B. It should be noted that the Condar method is based on
flue gas temperatures at approximately 1.5 meters above the stove,
while in this study, flue gas temperatures were measured at about
0.5 meters. Gas temperatures measured in this study are therefore
higher (estimated to be 40-100°C higher) than would be using the
Condar lab procedure. Higher flue gas temperatures result in lower
calculated thermal efficiencies.
Particulate material is also used as a measure of combustion
efficiency in the Condar system. Normally measured at 1.5 meters in
the lab, field measurements of particulate material were made about
0.5 meters above the stove. With potentially higher particle
loadings due to less flue pipe deposition, calculated efficiencies
may be lower from field values than would be observed in the lab
conditions used to calibrate the Condar method. For these reasons,
efficiency values for this project are thought to be artificially
low. Condar-calculated efficiencies are, however, used in
2-12
-------
calculating grams per million joule emission rates, which may result
in these values being artificially high relative to laboratory-
generated values, and should be used with great caution.
Calculation of emissions as a function of time (g/hr) required, in addition to "1"
through "5":
8. Mass of dry wood burned (measured by scale basket, corrected with
moisture measurements).
9. Total hours of stove operation. Data were computed on the basis of
hours of stove operation (stack temperature >100°F).
Particulate emission data were calculated for samples for which all necessary
parameters were currently available. These parameters include: lab particulate
weights from dichloromethane and methanol rinses, filter catch and resin (XAD-2)
extracts, valid field leak checks and flow calibrations, Data LOG'r data for
sampling duration and stove operation, valid oxygen calibrations, and valid scale
weighings.
COMBUSTOR LONGEVITY INSPECTIONS
Several efforts were made to evaluate the longevity of catalytic combustors in the
study and to assess combustor effectiveness over time. The original study design
included six catalytic stoves which had at least one heating season of use; these
"Group III" stoves were to be qualitatively compared with the catalytic stoves used
in Group I homes.
Based on reports of catalyst deterioration and preliminary emission performance
results from the first heating season, two tasks were added to the study for the
second heating season. These included:
Inspection of Catalytic Combustors
Between the first and second heating season, all catalytic combustors in the study
were removed and inspected. Combustors were evaluated visually for evidence of
plugging, cracking, erosion or structural damage, and peeling. Following the
second heating season, all available stoves and catalytic combustors used in the
study were inspected. Combustors were removed and replaced with new units. The
used combustors were archived for future testing. Results of the final stove
inspection will be reported under separate cover.
2-13
-------
Laboratory Testing of Field Combustors
Three combustors were returned to the laboratory after one heating season of use.
Combustors were selected based on participate emission rates measured during the
first heating season; combustors representing high, medium, and low emissions were
selected. All combustors were nominally six inches in diameter and three inches
thick (15 cm diameter, 7.6 cm thick), representing two manufacturers. Combustors
were tested in a Woodcutters Manufacturing "Blaze King Princess" prototype, which
had been used by the Oregon Department of Environmental Quality for certifying
combustors in its woodstove program. A single test run at an average burn rate of
about 1.2 kg/hr (dry) was conducted on each combustor to measure particulate
emission rates.
2-14
-------
Section 3
RESULTS AND DISCUSSION
CREOSOTE
Creosote deposition and removal is dependent on several interrelated factors.
Significant influences on creosote deposition that were documented include chimney
system construction, exposure, geometry, and height; woodstove technology type;
total weight of creosote collected; mean flue gas temperature; average burn rate;
and woodstove operational time. The most significant undocumented factor was
volatilization of creosote during high burn periods. Creosote mass in the chimney
can vary on a continual basis, as illustrated by the significant differences in
creosote deposition rates observed in some homes where creosote data was obtained
during two heating seasons.
Caution should be used in interpreting the creosote deposition rate data due to the
inherent difficulties in quantifying creosote accumulation. The creosote
accumulation values are intended only to serve as a general indication of creosote
accumulation. Further evaluation of the stove and flue system is recommended
before conclusions are established.
Stove Technology
Table 3-1 presents the creosote accumulations measured in individual study homes
for the 1985-1986 and 1987-1987 heating seasons. A summary of sample population,
mean, standard deviation, and maximum and minimum values by woodstove technology
type is also presented. The weight of creosote collected (total mass of material
removed from the chimney by sweeping) was normalized using heating degree-day data
(Fahrenheit scale) to give a creosote deposition rate in units of kilograms per
thousand heating degree-days (kg/1000 HDD). Only results with a high degree of
confidence were used; "atypical" results are presented (with explanations), but are
not included in data summaries or figures. Figure 3-1 presents the overall mean
creosote accumulations by stove technology type.
The conventional stoves had the highest overall mean creosote deposition rate (1.09
kg/1000 HDD), the highest maximum value (5.78 kg/1000 HDD), and the widest range of
3-1
-------
Table 3-1
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
V01-85/86
V01-86/87
V02-85/86
V02-86/87
V03-85/86
V03-86/87
V04-85/86
V04-86/87
V05-85/86
V05-86/87
V06-85/86
V06-86/87
V07-85/86
V07-86/87
V08-85/86
V08-86/87
V09-85/86
V10-85/86
V10-86/87
Vll-85/86
Vll-86/87
V12-85/86
V12-86/87
V13-85/86
V13-86/87
V14-85/86
V14-86/87
CREOSOTE ACCUMULATION (Kg/1000 HDD)3/
Catalytic
(3.28)b/c/
1.036/d/
0.30
0.90
0.49
0.79
0.82
0.06
0.31
0.39
Add-On/Retrofit
0.37
0.61
0.93
0.49
0.66
0.41
0.49
0.73b/
Low-Emission
0.29
0.16
0.11
0.72
(2.72)d/e/
Conventional
0.47
0.54
0.07
0.86b/
(Continued)
3-2
-------
Table 3-1 (Continued)
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
V15-85/86
V15-86/87
V16-85/86
V16-86/87
V17-85/86
V17-86/87
V18-85/86
V18-86/87
V19-85/86
V19-86/87
V20-85/86
V20-86/87
V21-85/86
V21-86/87
V22-85/86
V22-86/87
V23-85/86
V23-86/87
V24-85/86
V24-86/87
V25-85/86
V26-85/86
V26-86/87
V27-85/86
V27-86/87
V28-85/86
V28-86/87
CREOSOTE ACCUMULATION (Kg/1000 HDD)3/
Catalytic
0.68
0.90
0.26
0.24
0.70f/
0.95
0.66
0.38
0.37
0.35
Add-On/Retrofit
1.75
0.35
1.79
Low-Emission
0.15
1.08
Conventional
(3.32)e/
1.80b/
0.82
0.47
0.79
0.65
1.03
0.18f/
0.52
0.48
0.74
0.71b/
(Continued)
3-3
-------
Table 3-1 (Continued)
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
V29-85/86
V29-86/87
V30-85/86
V30-86/87
V31-85/86
V31-86/87
V32-85/86
V32-86/87
V33-85/86
V33-86/87
V34-86/87
V35-86/87
N01-85/86
N01-86/87
N02-85/86
N02-86/87
N03-85/86
N03-86/87
N04-85/86
N04-86/87
N05-85/86
N05-86/87
N06-85/86
N07-85/86
N07-86/87
CREOSOTE ACCUMULATION (Kg/1000 HDD)a/
Catalytic
0.27
0.35
0.59
1.98
0.279/
0.55
0.50
0.72
°-45H/
1.01d/
0.20
0.38
Add-On/Retrofit
0.60
0.33
0.79
2.90f/
3.59
0.46
Low-Emission
0.22f/
0.24
1.13
1.07
Conventional
0.72
0.72
0.33
(Continued)
3-4
-------
Table 3-1 (Continued)
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
N08-85/86
N08-86/87
N09-85/86
N09-86/87
N10-85/86
N10-86/87
Nll-85/86
Nll-86/87
N12-85/86
N12-86/87
N13-85/86
N13-86/87
N14-85/86
N14-86/87
N15-85/86
N15-86/87
N16-85/86
N16-86/87
N17-85/86
N17-86/87
N18-85/86
N18-86/87
N19-85/86
N20-85/86
N20-86/87
N21-85/86
CREOSOTE ACCUMULATION (Kg/1000 HDD)3/
Catalytic
0.82
1.43
0.86
1.43
0.60
0.39
0.54
0.06
0.99
Add-On/Retrofit
0.14
0.20f/
1.69
1.64
0.33f/
Low-Emission
1.05
0.36
0.24
0.33
Conventional
1.50
5.78
2.05
1.14
0.61
1.42
2.37
1.37
(Continued)
3-5
-------
Table 3-1 (Continued)
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
N22-85/86
N22-86/87
N23-85/86
N24-85/86
N24-86/87
N25-85/86
N25-86/87
N26-85/86
N26-86/87
N27-85/86
N27-86/87
N28-85/86
N29-85/86
N29-86/87
N30-85/86
N31-85/86
N31-86/87
N32-85/86
N32-86/87
N33-85/86
N33-86/87
CREOSOTE ACCUMULATION (Kg/1000 HDD)3/
Catalytic
1.09
0.62
0.30
0.13
0.04
0.14
0.93
0.80
Add-On/Retrofit
0.52
0.36
2.57
Low-Emission
0.74
Conventional
0.86
0,78
0.95
1.02
0.73
2.06
1.31
0.97
1.31
(Continued)
3-6
-------
Table 3-1 (Continued)
CREOSOTE ACCUMULATION BY WOODSTOVE TECHNOLOGY TYPE
STUDY HOME
AND
HEATING SEASON
Sample
Population
Mean
Standard
Deviation (a]
Minimum
Value
Maximum
Value
CREOSOTE ACCUMULATION (Kg/1000 HDD)3/
Catalytic
48 (49)c/
0.60(0.66)
0.40(0.54)
0.04
1.98(3.28)
Add-On/Retrofit
25
0.99
0.90
0.14
3.59
Low-Emission
15 (16)b/
0.53(0.66)
0.38(0.65)
0.11
1.13(2.72)
Conventional
35 (36)b/
1.09(1.15)
0.96(1.02)
0.07
5.78(5.78)
a/ Values inside parentheses indicate data which may be atypical or non-
representative. See notes for specific details. Data summaries were calculated
without "atypical" data (no parentheses) and including these data (in
parentheses).
b/ Additional unscheduled chimney cleaning was performed by homeowner or fire
department in which a small portion of the season's total creosote may not have
been recorded.
c/ ( ) denotes values obtained using V05-85/86 data. V05 combustor was jarred
out of position during installation of stove and corrected in the spring of 1986.
Values not believed to be typical of catalytic technology.
d/ The collected creosote included a significant quantity of water or ice which
was subtracted from the total chimney sweeping sample weight.
e/ ( ) denotes values obtained using V14-86/87 and V15-86/87 data. Homes V14 and
V15 had a power brush chimney cleaning in early 1987 to remove a thick creosote
glaze from the chimney. Values are believed to include chimney deposits from
several years of developing this creosote glaze.
f/ Creosote accumulation results are from a relatively short time period (50-75
days of the heating season).
9/ HDD for October 22-30, 1985 estimated at 160 total for period.
3-7
-------
CO
1.2 -
i.O -
MEAN
CREOSOTE
ACCUMULATION
(KG/1000 HOD)
0.6 -
O.H -
0.2 -
Figure 3-1
Creosote Accumulation by Stove Technology
1
CflTftLVTIC
STDUES
ADU-DN/
RETROFITS
LOU
EHISSIDH
STDUES
CDHUENTIDNAL
STOUES
-------
observed values (0.07 kg/1000 HDD to 5.78 kg/1000 HDD). The low-emission stoves
had the lowest overall mean creosote deposition rate (0.53 kg/1000 HDD). The mean
creosote deposition rate for the add-on/retrofit devices (0.99 kg/1000 HDD) was
slightly less than the mean creosote deposition rate for the conventional stoves.
The overall mean creosote deposition rate for the catalytic stoves (0.60 kg/1000
HDD) was similar to the mean creosote deposition rate for the low-emission stoves.
Caution should be used in comparing the mean creosote deposition rates by stove
technology due to the considerable influence of individual chimney systems on
creosote deposition rate. Further examination of these factors is presented in
Section 4.
Chimney System
Table 3-2 presents data on creosote accumulation by three general chimney system
types; round prefabricated metal, rectangular masonry with indoor exposure, and
rectangular masonry with outdoor exposure. Data from a fourth category (chimney
systems which do not fit into the above three classifications) are also included in
Table 3-2. Figure 3-2 presents the overall mean creosote accumulations by chimney
configuration.
The round prefabricated metal chimney systems had the lowest mean creosote
deposition rate (0.41 kg/1000 HDD), the masonry chimneys located inside the
exterior walls of the house had an intermediate mean creosote deposition rate
(0.68 kg/1000 HDD), and the masonry chimneys located outside the exterior walls of
the house had the highest mean creosote deposition rate (1.14 kg/1000 HDD).
Caution should be used when comparing overall mean creosote accumulations due to
different populations of woodstove technologies used with similar chimneys. The
mixture of woodstove technologies was different for each chimney type, so the
overall mean creosote accumulations by chimney type should be considered only an
indication of relative creosote accumulations.
The ranking of creosote accumulation by chimney type was generally as would be
expected. The round prefabricated metal chimneys were all less than or equal to
20.3 cm (eight inches) in diameter, which would be expected to produce higher flue
gas temperatures and better draft conditions than masonry chimneys with larger
cross-sectional area. Better draft generally results in less creosote deposition
due to less cooling of flue gases.
3-9
-------
Table 3-2
CREOSOTE ACCUMULATION BY CHIMNEY CONFIGURATION
PARAMETER
Sample Population
(Total)
Catalytic Stove
Add-on/Retrofit Stove
Low-Emission Stove
Conventional Stove
Mean
Standard Deviation(cr)
Minimum Value
Maximum Value
CREOSOTE ACCUMULATION (Kg/1000 HDD)
Round Prefab.
Metal Chimney3/
27
11
4
5
7
0.41
0.27
0.04
1.09
Interior
Masonry Chimney"'
35
14
4
6
11
0.68
0.42
0.06
1.98
Exterior
Masonry Chimney0'
50 (53)d/
18 (19)
17 (17)
2 (3)
13 (14)
1.14 (1.25)6/
0.99 (1.06)6/
0.13 (0.13)
5.78 (5.78)
a/ Round prefabricated metal chimneys with six-inch, seven-inch, or eight-inch
inside diameters (chimney types I, II, III, and XI in home characteristics
table).
"I Rectangular tile-lined masonry chimneys located inside the exterior walls of
the house with 7" x 7" or 7" x 11" flue cross-section sizes (chimney types V and
VI in home characteristics table).
c' Rectangular tile-lined masonry chimneys located outside the exterior walls of
the house with 7" x 7" or 7" x 11" flue cross-section sizes (chimney types VII,
VIII, and IX in home characteristics table).
d/ Values inside parentheses () include data from V05-85/86, V14-86/87, and V15-
86/87. These homes had either a misaligned combustor (V05) or a power brush
chimney cleaning (V14 and V15) which was likely responsible for higher creosote
accumulation values.
e' Note that the Exterior Masonry Chimney category has a higher proportion of
Add-on/Retrofit Stoves. High particulate emissions from these devices,
documented in Table 6, may contribute to the higher mean value for this chimney
category.
3-10
-------
Figure 3-2
Creosote Accumulation by Chimney Configuration
CO
I
1.2 -
1,0 -I
HEftN
CREOSOTE
ftCCUHULflTIQN
(KG/1000 H:DD)
0.5 -
0.1 -
0.2 -
HETftL
PREFRBRICflTED
CHIMNEY
HflSDNRY
CIHTERIDR)
CHIMNEV
HftSDNRV
(EKTERIDR)
CHIMNEV
-------
Some of the masonry chimney systems had cross-sectional areas of up to 17.8 cm by
27.9 cm (7 inches by 11 inches). These larger chimneys would be expected to
produce poorer draft conditions relative to the smaller round prefabricated metal
chimney systems; consequently, the mean creosote deposition rate for the masonry
chimneys would be expected to be higher than the mean rate for the metal chimneys
due to increased flue gas cooling and poorer draft. The interior masonry chimneys
were not exposed to cold outside temperatures (and potentially increased heat
transfer from the flue gas stream), so flue gas temperatures in the interior
masonry chimneys would be relatively higher than in the exterior masonry chimneys.
Individual Installations
Table 3-3 presents creosote accumulation data by home for individual woodstove
models, and lists the chimney type for each home. Caution should be used in
comparing the mean creosote deposition rates of individual stove models due to the
considerable influence of the individual chimney systems on creosote deposition
rates.
Catalytic Stoves. The range of mean creosote deposition rates for catalytic stoves
was 0.53 kg/1000 HDD (Stove Model C) to 0.79 kg/1000 HDD (Stove Model B). The
range of values is relatively narrow given the inherent difficulties in quantifying
creosote deposition rates and the mixture of chimney types within the individual
catalytic stove model data sets. The data from Stove Model P (catalytic stoves in
place prior to the start of the study) is presented for individual homes without a
mean, as this category consists of six different stove models.
Add-on/Retrofits. The range of mean creosote deposition rates for the add-on/
retrofit devices was 0.49 kg/1000 HDD (Stove Model E) to 1.66 kg/1000 HDD (Stove
Model F). The add-on/retrofit category as a group contained a fairly high
percentage of exterior masonry chimneys, which may have contributed to a high bias
of the overall mean creosote deposition rate for this technology classification
(0.99 kg/1000 HDD). Creosote accumulation data for the add-on/retrofit category
included four creosote samples from prefabricated metal chimneys (16% of the total
data set), four samples from interior masonry chimneys (16% of the total data set),
and 17 samples from exterior masonry chimneys (68% of the total data set). For the
other technology classifications, the exterior masonry chimney percentages were
lower than for the add-on/retrofit classification (39% for catalytic stoves, 19%
for low-emission stoves, and 39% for conventional stoves).
3-12
-------
Table 3-3
CREOSOTE ACCUMULATION BY STOVE MODEL
CATALYTIC STOVES
Stove
Code
A
B
C
Home
Code
V22
V28
N01
N10
N20
AVERAGE
V05
Vll
V20
N09
N18
N22
N23
AVERAGE
V07
V16
V19
V26
N03
N19
AVERAGE
Heating
Season
86/87
86/87
85/86
86/87
85/86
86/87
86/87
86/87
85/86
86/87
86/87
85/86
86/87
86/87
86/87
85/86
85/86
86/87
85/86
86/87
86/87
85/86
85/86
86/87
85/86
Creosote Accumulation
(kg/1000 HDD)
0.38
0.35
0.50
0.72
0.86
1.43
0.99
0.75
1.03b/c/
0.82
0.06
0.66
0.82
1.43
0.54
1.09
0.62
0.79
0.30
0.90
0.68
0.90
0.95
0.37
0.20
0.38
0.06
0.53
Chimney System
Classification3'
b
d
a
a
c
c
c
c
b
d
a
c
c
c.
a
c
c
c
c
c
b
d
b
b
a
(Continued)
3-13
-------
Table 3-3 (Continued)
CATALYTIC STOVES (Continued)
Stove
Code
D
P
Home
Code
V08
V13
V18
N02
Nil
AVERAGE
V17
V31
V32
V33
N31
N32
N33
AVERAGE
Heating
Season
85/86
86/87
85/86
86/87
85/86
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
Creosote Accumulation
(kg/1000 HDD)
0.49
0.79
0.31
0.39..
0.70d/
0.45 ,
1.01C/
0.60
0.39
0.57
0.26
0.24
0.27
0.35
0.59
1.98
0.276/
0.55
0.30
0.13
0.04
1.14
0.93
0.80
0.56
Chimney System
Classification3/
b
b
b
b
b
c
c
b
b
a
a
a
a
b
b
c
c
c
c
a
a
d
d
(Continued)
3-14
-------
Table 3-3 (Continued)
ADD-ON/RETROFITS
Stove
Code
E
F
G
H
I
J
Home
Code
V01
V29
N24
N26
AVERAGE
V03
V12
N05
AVERAGE
V02
V21
N04
N27
AVERAGE
V10
V15
N13
AVERAGE
V24
V30
N06
N12
N14
AVERAGE
V10
N04
N12
N14
AVERAGE
Heating
Season
85/86
86/87
86/87
86/87
86/87
85/86
85/86
85/86
85/86
86/87
86/87
85/86
86/87
85/86
85/86
85/86
86/87
86/87
85/86
85/86
85/86
86/87
86/87
86/87
86/87
Creosote Accumulation
v (kg/1000 HDD)
0.37
0.61
0.60
0.52
0.36
0.49
0.66
0.73b/
3.59
1.66
0.93
0.49
0.35
0.79
2.57
1.03
0.41
1.75
1.69
1.28
1.79
0.33
0.46
0.14
1.64
0.87
°'49H/
2. god/
0.205/
0.33d/
0.98
Chimney System
Classification
c
c
b
a
b
b
b
c
c
c
a
c
c
c
c
c
c
c
c
a
c
c
c
a
c
(Continued)
3-15
-------
Table 3-3 (Continued)
LOW-EMISSION STOVES
Stove
Code
K
L
M
N
Home
Code
V18
V23
N07
N29
AVERAGE
V04
N15
AVERAGE
V12
V34
N13
AVERAGE
V03
V35
N16
AVERAGE
Heating
Season
86/87
86/87
85/86
86/87
86/87
85/86
86/87
85/86
86/87
86/87
86/87
86/87
86/87
86/87
86/87
Creosote Accumulation
(kg/1000 HDD)
0.15
1.08
1.13
1.07
0.74
0.83
0.16
0.11
0.36
0.24
0.22
0.72
0.22d/
1.05
0.66
0.29
0.24
0.33
0.29
Chimney System
Classification
a
b
d
d
b
b
b
a
a
b
a
c
b
a
c
(Continued)
3-16
-------
Table 3-3 (Continued)
CONVENTIONAL STOVES
Stove
Code
0
Home
Code
V06
V09
V14
V19
V20
V21
V22
V23
V24
V25
V26
V27
V28
V29
V30
N05
N08
N16
N17
N18
N20
N21
N22
N24
N25
N26
N27
N28
N29
N30
AVERAGE
Heating
Season
85/86
86/87
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
86/87
85/86
86/87
85/86
85/86
85/86
86/87
85/86
86/87
85/86
85/86
86/87
85/86
85/86
85/86
85/86
85/86
85/86
86/87
85/86
85/86
85/86
85/86
85/86
Creosote Accumulation
(kg/1000 HDD)
0.47
0.54
0.07U/
0.86b/
1.80b/
0.82
0.47
0.79
0.65
l-Q3j,
0.18d/
0.52
0.48
0.74u/
0.71b/
0.72
0.72
0.33
1.50
5.78
2.05
1.14
0.61
1.42
2.37
1.37
0.86
0.78
0.95
1.02
0.73
2.06
1.31
0.97
1.31
1.09
Chimney System
Classification
b
b
b
c
b
a
a
b
b
c
a
d
a
a
d
b
c
c
c
c
c
c
c
c
c
b
a
a
d
d
b
c
b
b
c
(Continued)
3-17
-------
Table 3-3 (Continued)
a/ Chimney systems are classified as follows:
a. Round prefabricated metal chimneys with six-inch, seven-inch, or eight-inch
inside diameters (chimney types I, II, III, and XI in home characteristics
table). Most of the "Type A" chimneys were interior installations; three
(N19, V17, N15) were exterior. Creosote accumulation in the three outside
installations was no higher than in the inside installations, so all "Type A"
chimneys were grouped together.
b. Rectangular tile-lined masonry chimneys located inside the exterior walls
of the house with 7" x 7" or 7" x 11" flue cross-section sizes (chimney types
V and VI in home characteristics table).
c. Rectangular tile-lined masonry chimneys located outside the exterior walls
of the house with 7" x 7" or 7" x 11" flue cross-section sizes (chimney types
VII, VIII, and IX in home characteristics table).
d. Chimneys that do not fit into above categories a, b, or c (chimney types
IV, X, XII, and XIII in home characteristics table). Includes the following
chimney types:
Vll, N33--Round tile-lined masonry chimney located inside the exterior
walls of the house (cross-sectional area 8 inches in diameter).
V26, V28--Stainless steel-lined masonry chimney located inside the exterior
walls of the house.
N07--Round tile-lined masonry chimney located outside the exterior walls of
the house (cross-sectional area 8 inches in diameter).
N25--Stove vents into a fireplace with a rectangular tile-lined masonry
chimney located outside the exterior walls of the house (cross-
sectional area approximately 7 inches by 11 inches).
"i Additional unscheduled chimney cleaning was performed by homeowner or fire
department. A small portion of the season's total creosote may not have been
recorded.
c/ The collected creosote included a significant quantity of water or ice which
was subtracted from the total chimney sweeping sample weight.
d/ Creosote accumulation results are from a relatively short time period (50-75
days of the heating season).
e/ HDD for October 22-30, 1985, estimated at 160 total for period.
3-18
-------
Low-emission Stoves. The range of mean creosote deposition rates for the low-
emission stove models was 0.22 kg/1000 HDD (Stove Model L) to 0.83 kg/1000 HDD
(Stove Model K). As previously indicated, the percentage of exterior masonry
chimneys in the low-emission technology classification was the lowest percentage of
all technology classifications (25%). This characteristic of the data set probably
resulted in a low bias of the creosote deposition rates in the low-emission
classification relative to other technology classifications.
Conventional Stoves. The range of creosote deposition rates for individual heating
seasons for the conventional stove models in individual homes was 0.07 kg/1000 HDD
(V09, 85/86) to 5.78 kg/1000 HDD (N08, 86/87). The overall mean creosote
deposition rate for the conventional stoves was 1.09 kg/1000 HDD. A total of 23
conventional stove models were evaluated in the study, each with unique design
characteristics. The significant differences in design, firebox size, and chimney
system in the conventional stove data set probably contributed to the relatively
wide range of observed creosote deposition rates.
Stove Switching
Table 3-4 presents creosote accumulation data from the 24 Group II homes where
woodstove technology switches were performed during the study. This approach was
intended to identify differences in creosote accumulation between stove types while
holding the stove operators, chimney system, and heating demand (house size)
constant. Comparisons of creosote accumulations are presented for catalytic stoves
vs. conventional stoves, add-on/retrofits vs. conventional stoves, low-emission
stoves vs. conventional stoves, and low-emission stoves vs. add-on/retrofits.
Figure 3-3 presents the mean percentage creosote accumulation decrease for
catalytic stoves, add-on/retrofits, and low-emission stoves versus conventional
stoves, and for low-emission stoves versus add-on/retrofits.
The catalytic stoves exhibited the largest mean percentage creosote deposition rate
decrease (37%) when switching from conventional stoves. The add-on/retrofits and
low-emission stoves had mean percentage creosote deposition rate decreases versus
conventional technology stoves that were within 2% (absolute) of each other (12%
decrease for add-on/retrofits, 14% decrease for low-emission stoves). It should be
noted that the data set for the low-emission vs. conventional stoves consisted of
three values (in comparison to eight values for catalytic vs. conventional stoves
and seven values for add-on/retrofits vs. conventional stoves). The low-emission
stoves showed a 32% decrease in creosote deposition rate versus the add-on/
retrofits. The data set for the low-emission stoves vs. add-on/retrofits was also
3-19
-------
Table 3-4
EFFECTS OF STOVE TECHNOLOGY CHANGES ON CREOSOTE ACCUMULATION9/13/
CATALYTIC VS. CONVENTIONAL:
STUDY HOME
V19
V20
V22
V26
V28
N18
N20
N22
Average
CONVENTIONAL STOVE
(Kg Creosote/1000 HDD)
1.80
0.82
0.79
0.52
0.71
1.42
2.37
0.86
1.16
CATALYTIC STOVE
(Kg Creosote/1000 HDD)
0.95
0.66
0.38
0.37
0.35
0.54
0.99
1.09
0.67
NET CHANGE IN CREOSOTE
ACCUMULATION (%)
-47
-20
-52
-29
-51
-62
-58
+27
-37 [27]
ADD-ON/RETROFIT VS. CONVENTIONAL:
STUDY HOME
V15
V21
V24
V29
V30
N05
N24
N26
N27
Average
CONVENTIONAL STOVE
(Kg Creosote/1000 HDD)
(3.32)c/
0.47
1.03
0.72
0.72
(0.33)
0.78
0.73
2.06
0.93 (1.13)
ADD-ON/RETOFIT
(Kg Creosote/1000 HDD)
(1-75)
0.35
1.79
0.60
0.33
(3.59)
0.52
0.36
2.57
0.93 (1.32)
NET CHANGE IN CREOSOTE
ACCUMULATION (%)
(-47)
-26
+74
-17
-54
(+988)d/
-33
-51
+25
-12 [43]
(+95 [318])
LOW-EMISSION VS. CONVENTIONAL:
STUDY HOME
V14
V23
N16
N29
Average
CONVENTIONAL STOVE
(Kg Creosote/1000 HDD)
(0.86)
0.65
2.05
0.97
1.22 (1.13)
LOW-EMISSION STOVE
(Kg Creosote/1000 HDD)
(2.72)c/
1.08
0.33
0.74
0.72 (1.22)
NET CHANGE IN CREOSOTE
ACCUMULATION (%)
(+216)
+66
-84
-24
-14 [62]
(+43.5 [113])
(Continued)
3-20
-------
Table 3-4 (Continued)
EFFECTS OF STOVE TECHNOLOGY CHANGES ON CREOSOTE ACCUMULATION
LOW-EMISSION VS. ADD-ON/RETROFIT:
STUDY HOME
V03
V12
N13
Average
ADD-ON/RETROFIT
(Kg Creosote/1000 HDD)
0.66
0.73
1.69
1.03
LOW-EMISSION STOVE
(Kg Creosote/1000 HDD)
0.29
0.72
1.05
0.69
NET CHANGE IN CREOSOTE
ACCUMULATION (%)
-56
-1
-38
-32 [23]
a' Values in brackets [ ] are standard deviations (or).
b/ Values in parentheses () indicate data which may be atypical or
unrepresentative. Data summaries are calculated without atypical data (no
parentheses) and including these data (in parentheses). See notes for specific
details.
c/ Homes V14 and V15 had a power brush chimney cleaning in early 1987 to remove a
thick creosote glaze from the chimney. These creosote values may include some
chimney deposits from several years of developing this creosote glaze.
d/ Data from home V05 exhibits creosote accumulation with retrofit technology at
over ten times the creosote accumulation with a conventional stove. Because of
this unusually large differential and the fact that this retrofit model was
discontinued in 1986, data from this home are listed as "atypical data."
3-21
-------
50
CDHPftRflTIUE
CREDSDTE
REDUCTION
(X)
20 -
10 -
Figure 3-3
Comparative Creosote Accumulation.' Group II Homes
VAVAW
•X^XOX*
*x*x*x*x*
tVAV+V+V
CA1ALVTIC JTDUES
US.
CDHUENTIDNftL STDUES
(8 HOMES)
AOD-DN/REIRDFITS
US.
CDNUENTIQNflL STOUES
(7 HOMES)
LOW EMISSION STDUES LDH
US
COHUEHTIDNflL STDUES
(3 HOMES)
EMISSION STDUES
US .
flDD-DN/RETRDFITS
(3 HOMES)
-------
relatively small (three values). The consistent reductions exhibited by the
advanced technology stoves indicates that these stoves do reduce the amount of
creosote deposited in the chimney system.
WOOD USE
Stove Technology (Scale Weighings)
Table 3-5 presents average wood use data compiled by individual home and
categorized by stove technology groupings. Data were determined from wood weight
data recorded by the Data LOG'r scale units as stoves were fueled. The total mass
of wood burned during each one-week sampling period was normalized by heating
degree-days (Fahrenheit basis). Other factors, such as the size of the house
heated and the use frequency of heating sources other than the woodstove are also
tabulated. Scale weighing values are not reported if improper use of the scale was
reported by the homeowner or values were "suspect" based on data file reviews.
Figure 3-4 presents the mean wood use (kg/HDD) by stove technology type (based on
scale weighing measurements).
High variability of wood use data was expected due to several factors. The amount
of time the stove was operated is a significant factor and wood use data tends to
favor stoves which are not operated on a continuous basis; stoves which were
allowed to burn out (cold-to-cold burn cycle) would have less total burning time
than the stoves that were operated continuously. A long burn-out "tail" will
result in a lower average burn rate.
The area (or volume) of the home being heated by the woodstove and the frequency of
use of other heating sources can also significantly affect wood usage. "Heated
area" can be difficult to quantify, as convective heat distribution through a house
may be highly variable. Areas in the home being heated are thought to vary
diurnally and by weekdays vs. weekends. Different temperatures were maintained in
individual homes, requiring different heat inputs. Fuel usage from appliances
other than the woodstove can be calculated from oil, gas, and electric bills, but
may be questionable if appliances other than heaters are served by the same energy
source. Instrumenting individual appliances was beyond the scope of this project.
The lowest overall average wood use value measured by scale weighings (0.53 kg dry
wood/HDD) was obtained with the low-emission stoves. The catalytic stove group had
the second lowest average wood use value (0.64 kg dry wood/HDD). The add-on/
retrofit group and the conventional stove group showed similar average wood use
3-23
-------
Table 3-5
WOOD USE -- SCALE WEIGHING MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
V01-85/86
V01-86/87
V02-85/86
V03-85/86
V03-86/87
V04-85/86
V04-86/87
V05-85/86
V05-86/87
V06-85/86
V06-86/87
V07-85/86
V07-86/87
V08-85/86
V08-86/87
V09-85/86
V10-85/86
V10-86/87
Vll-85/86
Vll-86/87
V12-85/86
V12-86/87
V13-85/86
V13-86/87
V14-85/86
V14-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.65
0.51
0.86
0.60
0.60
0.45
0.46
0.46
0.55
0.65
Add-on/
Retrofit
0.76
0.75
0.78
0.67
0.44
0.40
0.52
Low-
Emission
0.69
0.49
0.33
0.36
0.46
Conventional
0.96
0.86
0.45
0.72
Heated3/
Area
MS
MS
MS
MS
ML
ML
ML
S
S
ML
MS
MS
MS
S
Frequency of
Alt. Heat Use
Never
Never
Frequently
Frequently
Occasionally
Rarely
Occasionally
Rarely
Rarely
Occasionally
Rarely
Rarely
Rarely
Frequently
(Continued)
3-24
-------
Table 3-5 (Continued)
WOOD USE -- SCALE WEIGHING MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
V16-85/86
V16-86/87
V18-86/87
V31-85/86
V31-86/87
V32-85/86
V32-86/87
V34-86/87
V35-86/87
N01-85/86
N01-86/87
N02-85/86
N02-86/87
N03-85/86
N03-86/87
N04-85/86
N04-86/87
N05-85/86
N05-86/87
N06-85/86
N07-85/86
N07-86/87
N08-85/86
N08-86/87
N09-85/86
N09-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.60
0.51
0.52
0.57
0.91
0.47
0.53
0.49
0.85
0.49
0.47
0.40
0.58
0.75
Add-on/
Retrofit
1.08
1.08
0.61
1.47
Low-
Emission
0.55
0.50
0.42
0.71
0.57
Conventional
0.51
1.58
1.54
Heated3/
Area
ML
S
S
L
ML
MS
L
L
L
L
ML
L
ML
ML
L
Frequency of
Alt. Heat Use
Rarely
Occasionally
Frequently
Rarely
Frequently
Occasionally
Rarely
Rarely
Frequently
Occasionally
Frequently
Occasionally
Occasional ly
Occasionally
Occasionally
(Continued)
3-25
-------
Table 3-5 (Continued)
WOOD USE -- SCALE WEIGHING MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
N10-85/86
N10-86/87
Nll-85/86
Nll-86/87
N12-85/86
N12-86/87
N13-85/86
N13-86/87
N14-85/86
N14-86/87
N15-85/86
N15-86/87
N16-85/86
N16-86/87
N18-86/87
N31-85/86
N32-85/86
N32-86/87
N33-85/86
N33-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.74
0.83
0.41
0.44
0.93
0.51
0.70
0.65
1.21
1.15
Add-on/
Retrofit
0.92
0.63
0.91
1.50
1.09
Low-
Emission
0.55
0.57
0.62
0.56
Conventional
0.68
Heated3/
Area
L
ML
ML
L
L
ML
MS
MS
ML
ML
L
Frequency of
AH. Heat Use
Daily
Occasional ly
Frequently
Rarely
Rarely
Daily
Frequently
Daily
Never
Rarely
Never
(Continued)
3-26
-------
Table 3-5 (Continued)
WOOD USE -- SCALE WEIGHING MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
Sample
Population
Mean
Standard
Deviation (a)
Minimum
Value
Maximum
Value
WOOD USE (Dry Kg/HDD)
Catalytic
34
0.64
0.20
0.40
1.21
Add-on/
Retrofit
16
0.85
0.32
0.40
1.50
Low-
Emission
14
0.53
0.11
0.33
0.71
Conventional
8
0.91
0.40
0.45
1.58
Heated3/
Area
Frequency of
Alt. Heat Use
a/ House Size:
S = < 1000 ft2 heated
MS = >1000 < 1500 ft2 heated
ML = >1500 < 2000 ft2 heated
L = >2000 ft2 heated
3-27
-------
HEflN
UDDD USE
(DRV KG/HDD)
I
ro
Co
Figure 3-4
Wood Use by Stove Technology (Scale Weighing Measurements)
1.2 -
i.O -
0,6 -
0.4-
0.2 -
CftTftLVTIC
STDUES
ADD-DN/
RETROFITS
LDH
EMISSION
STDUES
COHUENTIDHAL
STDUES
-------
values (0.85 kg dry wood/HDD for the add-on/retrofits, 0.91 kg dry wood/HDD for the
conventional stoves).
Stove Technology (Woodpile Measurements)
Table 3-6 presents data similar to that presented in Table 3-5; however, Table 3-6
data are based on measurements of the woodpiles at each individual home. The
number of woodpile wood use values (103) is greater than the number of scale
weighing wood use values (72), as data from the Group II (uninstrumented) homes had
woodpile, but no scale, measurements. Woodpile measurements were conducted for
three purposes:
1. To document wood use in homes without expensive and intrusive
instrumentation.
2. To compare wood use measured by scale weighing with woodpile
measurements to assess the accuracy of the less-complex woodpile
measurements for future studies.
3. To provide a qualitative assessment of relative stove efficiency by
switching stove technologies between the 1985-86.and 1986-87 heating
seasons.
Woodpile measurements were recognized as having inherent inaccuracies due to
differences in stacking densities, determination of fuel species and densities,
estimating percentage mixtures of fuel species in individual woodpiles, making
accurate measurements of complex woodpile shapes, and undocumented removal or
additions to woodpiles. When serious problems such as those mentioned above were
noted, data were not reported.
The lowest overall average woodpile wood use value (0.46 kg dry wood/HDD) was
obtained with the low-emission stoves. The catalytic stove group had the second
lowest average wood use value (0.67 kg dry wood/HDD). As in the case of the scale
weighing wood use measurements, the add-on/retrofit group and the conventional
stove group showed similar average wood use values (0.85 kg dry wood/HDD for the
add-on/retrofits, 0.89 kg dry wood/HDD for the conventional stoves).
Method Comparisons
Table 3-7 presents a comparison of wood use data based on scale weighings and
woodpile measurements. Although average wood use values determined by the two
measurement methods were observed to vary significantly for individual homes, the
agreement between means and ranges was remarkably good. While comparisons include
different sample populations sizes (Group II homes with woodpile measurements did
3-29
-------
Table 3-6
WOOD USE -- WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
V01-85/86
V01-86/87
V02-85/86
V02-86/87
V03-85/86
V03-86/87
V04-85/86
V04-86/87
V05-85/86
V05-86/87
V06-85/86
V06-86/87
V07-85/86
V07-86/87
V08-85/86
V08-86/87
V09-85/86
V10-85/86
V10-86/87
Vll-85/86
Vll-86/87
V12-85/86
V12-86/87
V13-85/86
V13-86/87
V14-85/86
V14-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.62
0.55
0.97
0.80
0.68
0.63
0.61
0.35
0.87
0.65
Add-On/Retrofit
0.85
0.64
0.99
0.56
0.62
0.38
0.22
0.53
Low-Emission
0.59
0.36
0.29
0.60
M
Conventional
M
M
M
0.82
(Continued)
3-30
-------
Table 3-6 (Continued)
WOOD USE -- WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
V15-85/86
V15-86/87
V16-85/86
V16-86/87
V17-85/86
V17-86/87
V18-85/86
V18-86/87
V19-85/86
V19-86/87
V20-85/86
V20-86/87
V21-85/86
V21-86/87
V22-85/86
V22-86/87
V23-85/86
V23-86/87
V24-85/86
V24-86/87
V25-85/86
V26-85/86
V26-86/87
V27-85/86
V27-86/87
V28-85/86
V28-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.71
0.51
M
0.44
M
0.55
0.95
0.57
M
0.38
Add-On/Retrofit
0.60
0.37
1.57
Low-Emission
M
0.36
Conventional
0.79
1.22
1.16
0.53
0.80
0.51
M
M
0.40
M
0.57
0.54
(Continued)
3-31
-------
Table 3-6 (Continued)
WOOD USE -- WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
V29-85/86
V29-86/87
V30-85/86
V30-86/87
V31-85/86
V31-86/87
V32-85/86
V32-86/87
V33-85/86
V33-86/87
V34-86/87
V35-86/87
N01-85/86
N01-86/87
N02-85/86
N02-86/87
N03-85/86
N03-86/87
N04-85/86
N04-86/87
N05-85/86
N05-86/87
N06-85/86
N07-85/86
N07-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.66
0.60
0.81
0.62
0.72
0.57
0.50
0.54
0.86
0.94
0.31
0.35
Add-On/Retrofit
0.55
M
1.70
1.00
0.66
0.67
Low-Emission
0.59
0.51
0.67
M
Conventional
1.09
M
0.43
(Continued)
3-32
-------
Table 3-6 (Continued)
WOOD USE — WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
N08-85/86
N08-86/87
N09-85/86
N09-86/87
N10-85/86
N10-86/87
Nll-85/86
Nll-86/87
N12-85/86
N12-86/87
N13-85/86
N13-86/87
N14-85/86
N14-86/87
N15-85/86
N15-86/87
N16-85/86
N16-86/87
N17-85/86
N17-86/87
N18-85/86
N18-86/87
N19-85/86
N20-85/86
N20-86/87
N21-85/86
WOOD USE (Dry Kg/HDD)
Catalytic
M
0.71
0.68
0.74
0.58
0.67
0.59
0.82
0.53
Add-On/Retrofit
0.56
M
1.23
1.68
1.02
Low-Emission
0.39
0.46
0.28
0.39
Conventional
2.05
1.87
0.60
0.81
M
0.63
0.91
M
(Continued)
3-33
-------
Table 3-6 (Continued)
WOOD USE -- WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
N22-85/86
N22-86/87
N23-85/86
N24-85/86
N24-86/87
N25-85/86
N25-86/87
N26-85/86
N26-86/87
N27-85/86
N27-86/87
N28-85/86
N29-85/86
N29-86/87
N30-85/86
N31-85/86
N31-86/87
N32-85/86
N32-86/87
N33-85/86
N33-86/87
WOOD USE (Dry Kg/HDD)
Catalytic
0.89
M
0.47
0.35
1.15
0.76
1.29
0.90
Add-On/Retrofit
0.83
M
1.30
Low-Emission
0.45
Conventional
1.25
M
0.50
0.28
M
1.47
1.00
M
1.20
(Continued)
3-34
-------
Table 3-6 (Continued)
WOOD USE — WOODPILE MEASUREMENTS
STUDY HOME
AND
HEATING SEASON
Sample
Population
Mean
Standard
Deviation (a)
Minimum
Value
Maximum
Value
WOOD USE (Dry Kg/HDD)
Catalytic
44
0.67
0.21
0.31
1.29
Add-On/Retrofit
22
0.85
0.42
0.37
1.80
Low-Emission
13
0.46
0.12
0.28
0.67
Conventional
24
0.89
0.44
0.28
2.05
Note: M represents data which is missing due to measurement problems such as
unstacked or poorly stacked woodpiles, wood added to woodpile during season
without measurements, wood supply used for more than the woodstove under study,
late entry to study, or other factors.
3-35
-------
Table 3-7
WOOD USE-SCALE WEIGHING AND WOODPILE
MEASUREMENTS BY TECHNOLOGY TYPE
Technology
Catalytic
Stoves
Add-On/
Retrofits
Low-Emission
Stoves
Conventional
Stoves
Wood Use (dry kg/1000 HDD)
Scale Weighing Measurements9'
Mean
0.64
0.85
0.53
0.91
^
0.20
0.33
0.11
0.43
Nd/
34
16
14
8
Range
0.40-1.21
0.40-1.50
0.33-0.71
0.45-1.58
Woodpile Measurements'3'
Mean
0.67
0.85
0.46
0.89
*nc/
0.21
0.43
0.13
0.45
Nd/
44
22
13
24
Range
0.31-1.29
0.37-1.80
0.28-0.67
0.28-2.05
a/ Mass of wood burned, measured by Data LOG'r scale, normalized with heating
degree day data (Fahrenheit basis). Data were not reported if improper use of
the scale was reported by the homeowner or suspected based on data file reviews,
b/ Estimated mass of wood burned based on measurement of woodpiles, normalized
with heating degree day data (Fahrenheit basis). This method is recognized to
have inherent inaccuracies due to differences in stacking densities,
determination of fuel species and densities, estimating percentage mixtures of
fuel species in individual woodpiles, making accurate measurements of complex
woodpile shapes, and undocumented removal or additions to woodpiles. Data were
not reported when serious problems such as those mentioned above were noted.
c' Standard deviation.
d/ Sample population.
3-36
-------
not have scale instrumentation), preliminary assessment suggests that the woodpile
measurement technique could be used as a relatively good assessment tool to
document actual wood use in homes if a large sample population is used and careful
measurements of fuel moisture content and documentation of fuel species are made.
Tables 3-8A through 3-80 present a compilation of wood use data (both scale
weighing and woodpile measurement basis) by stove model. Although the scale
weighing and woodpile measurement wood use values are presented side by side,
caution should be used in making comparisons of these values due to small data sets
and the potential imprecision of individual woodpile measurements.
Catalytic Stoves. The catalytic stoves had a relatively narrow range of overall
average wood use values as measured by both methods (0.56 to 0.65 [Stove D, Stove
A] kg dry wood/HDD for scale weighings, 0.56 to 0.74 [Stove A, Stove D] kg dry
wood/HDD for woodpile measurements). Note that Stove D had the lowest overall
average wood use for the scale weighing method, and the highest overall average
wood use for the woodpile measurement method. This is probably an artifact of the
narrow range of overall average wood use values combined with the potential
imprecision of the woodpile measurement method when applied to relatively small
data sets.
Add-on/Retrofits. The add-on/retrofits had a wider range of average wood use
values (0.60 to 1.30 [Retrofit F, Add-on I] kg dry wood/HDD for scale weighings,
0.60 to 1.12 [Retrofit F, Add-on I] kg dry wood/HDD for woodpile measurements).
The average wood use measurements in the add-on/retrofit technology category should
be interpreted with caution due to the significant differences in firebox size for
the conventional stoves on which the add-on/retrofit devices were installed, in
addition to the inherent difficulties in comparing wood use previously discussed.
Low-emission Stoves. Like the catalytic stoves, the low-emission stoves also had a
relatively narrow range of average wood use values (0.47 to 0.61 [Stoves L and M,
Stove K] kg dry wood/HDD for scale weighings, 0.41 to 0.53 [Stoves L and M, Stove
K] kg dry wood/HDD for woodpile measurements). Like catalytic Stove D, low-
emission Stove M had one of the two lowest overall average wood use values for the
scale weighing method and the highest overall average wood use for the woodpile
measurement method. Again, this is probably an artifact of the narrow range of
overall average wood use values combined with the potential imprecision of the
woodpile measurement method when applied to relatively small data sets.
3-37
-------
Table 3-8A
WOOD USE BY STOVE MODEL -- CATALYTIC STOVES
Stove
Code
A
B
C
D
Home
Code
V22
V28
N01
N10
N20
AVERAGE
V05
Vll
V20
N09
N18
N22
AVERAGE
V07
V16
V19
N03
N19
AVERAGE
V08
V13
N02
Nil
AVERAGE
Heating
Season
86/87
86/87
85/86
86/87
85/86
86/87
86/87
85/86
86/87
85/86
86/87
86/87
85/86
86/87
86/87
86/87
85/86
86/87
85/86
86/87
86/87
85/86
86/87
85/86
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
Wood Use (dry kg/HDD)a/
Scale Weighing
b/
b/
0.53
0.49
0.74
0.83
b/
0.65 (N=4)
0.65
0.51
0.46
0.46
t>/
0.58
0.75
0,93
b/
0.62 (N=7)
0.86
0.60
0.60
0,51
b/
0.47
0,40
b/
0.57 (N=6)
0.60
0.45
0.55
0.65
0.85
0.49
0.41
0.44
0.56 (N=8)
Woodpile Measurements
0.57
0.38
0.50
0.54
0.68
0.74
0.53
0.56 (N=7)
0.66
0.55
0.61
0.35
0.95
c/
0.71
0.59
0.89
0.66 (N=8)
0.97
0.80
0.71
0.51
0.55
0.31
0.35
0.82
0.63 (N=8)
0.68
0.63
0.87
0.65
0.86
0.94
0.58
0.67
0.74 (N=8)
(Continued)
3-38
-------
Table 3-8A (Continued)
WOOD USE BY STOVE MODEL -- CATALYTIC STOVES
Stove
Code
P
Home
Code
V17
V31
V32
V33
N31
N32
N33
AVERAGE
Heating
Season
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
85/86
86/87
Wood Use (dry kg/HDD)a/
Scale Weighing
b/
0.52
0.57
0.91
0,47
b/
b/
W
b/
0.70
0.65
1.21
1.15
0.74 (N-9)
Woodpile Measurements
0.44
0.66
0.60
0.81
0.62
0.72
0.57
0.47
0.35
1.15
0.76
1.15
0.76
0.70 (N=13)
3-39
-------
Table 3-8B
WOOD USE BY STOVE MODEL -- ADD-ON/RETROFITS
Stove
Code
E
F
G
H
I
J
J&Gd/
J&0d/
Home
Code
V01
V29
N24
AVERAGE
V03
V12
N05
AVERAGE
V02
V21
N04
N27
AVERAGE
V10
V15
N13
AVERAGE
V24
N06
N12
N14
AVERAGE
V10
N04
N12
N14
Heating
Season
85/86
86/87
86/87
86/87
85/86
85/86
85/86
85/86
86/87
86/87
85/86
86/87
85/86
85/86
85/86
86/87
85/86
85/86
85/86
86/87
86/87
86/87
86/87
Wood Use (dry kg/HDD)a/
Scale Weighing
0.76
V,5
b/
0.76 (N=2)
0.67
0.52
0.61
0.60 (N=3)
°6?8
b/
'b98
0.93 (N=2)
V
0.91
0.68 (N=2)
b/
1.47
0.92
1.50
1.30 (N=3)
0.40
1.08
0.63
1.09
Woodpile Measurements
0.85
0.64
0.55
0.83
0.72 (N=4)
0.62
0.53
0.66
0.60 (N=3)
0.99
0.56
0.37
1.70
1.30
0.98 (N=5)
0.38
0.60
1.23
0.74 (N=3)
1.57
0.67
0.56
1.68
1.12 (N=4)
0.22
1.00
1.02
3-40
-------
Table 3-8C
WOOD USE BY STOVE MODEL -- LOW-EMISSION STOVES
Stove
Code
K
L
M
N
Home
Code
V18
V23
N07
N29
AVERAGE
V04
V34
N13
AVERAGE
V12
V14
V34
N13
AVERAGE
V03
V35
N16
AVERAGE
Heating
Season
86/87
86/87
85/86
86/87
86/87
85/86
86/87
86/87
86/87
86/87
86/87
86/87
86/87
86/87
86/87
86/87
Wood Use (dry kg/HDD)3/
Scale Weighing
°6?s
0.71
°6?7
0.61 (N=3)
0.49
0.33
0.50
0.55
0.47 (N=4)
0.36
0.46
0.50
0.55
0.47 (N=4)
0.69
0.42
0.56
0.56 (N=3)
Woodpile Measurements
c/
0.36
o.p
0.45
0.49 (N=3)
0.36
0.29
0.59
0.39
0.41 (N=4)
0.60
c/
0.59
0.39
0.53 (N=3)
0.59
0.51
0.39
0.50 (N=3)
3-41
-------
Table 3-8D
WOOD USE BY STOVE MODEL -- CONVENTIONAL STOVES
Stove
Code
0
Home
Code
V06
V09
V14
V15
V19
V20
V21
V22
V23
V26
V27
V28
V29
N05
N08
N16
N17
N18
N20
N22
N25
N27
N28
N30
AVERAGE
Heating
Season
85/86
86/87
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
86/87
86/87
85/86
85/86
86/87
85/86
86/87
85/86
85/86
85/86
85/86
85/86
85/86
86/87
85/86
85/86
85/86
Wood Use (dry kg/HDD)3/
Scale Weighing
0.96
0.86
0.45
0,72
b/
b/
i /
b/
b/
b/
b/
b/
b/
b/
b/
0.51
1.58
1.54
0,68
b/
b/
b/
b/
b/
b/
b/
b/
b/
0.91 (N=8)
Woodpile Measurements
c/
r- I
c/
^ /
c/
0.82
0.79
1.22
1.16
0.53
0.80
0.51
0.40
0.57
0.54
1.09
0.43
2.05
1.87
0.60
0.81
0.63
0.91
1.25
0.50
0.28
1.47
1.00
1.20
0.89 (N=24)
a> Wood use (dry kg/HDD) is presented for each heating season, home, and stove
model. Where applicable, two wood use measurement methods (scale measurements
and woodpile measurements) are presented. Although the two wood use measurement
methods are presented side-by-side, caution should be used in comparing results
for individual homes or stove models due to significant differences in
measurement methods and sample populations.
"' Group II home; not instrumented.
c' Data is missing due to woodpile measurement problems such as unstacked or
poorly stacked woodpiles, wood added to woodpile during season without
measurements, wood supply used for more than the study woodstove, late entry to
study, or other factors.
d/ Change in technology occurred during 86/87 heating season. Wood use
measurements represent a combined total for the two technologies.
3-42
-------
Conventional. The conventional stoves were not separated by stove model. Twenty-
three individual conventional stove models are represented in the data set, which
account for the relatively wide range of measured wood use values observed (0.45 to
1.58 kg dry wood/HDD for scale weighings, 0.28 to 2.05 kg dry wood/HDD for woodpile
measurements).
Stove Switching. Table 3-9 presents the results of switching stove technology in
Group II homes on wood use. Wood use, as measured by woodpile measurements, was
compared for homes which changed stove model between heating seasons. Figure 3-5
shows the mean percentage wood use decrease (woodpile measurements) for catalytic
stoves, add-on/retrofits, and low-emission stoves versus conventional stoves, and
the mean percentage wood use decrease for low-emission stoves versus add-on
retrofits.
All seven Group II homes which changed from conventional to catalytic technology
showed a decrease in wood use, from an average of 0.93 kg dry wood/HDD to an
average of 0.64 kg dry wood/HDD. Four of five homes switching from conventional to
add-on retrofit technology showed decreases in wood use. The average wood use for
the conventional vs. add-on/retrofit technology category decreased from 0.86 kg dry
wood/HDD to 0.70 kg dry wood/HDD. Both (two) homes which switched from
conventional stoves to low-emission stoves showed an average wood use decrease from
0.56 kg dry wood/HDD to 0.38 kg dry wood/HDD. Two of the three homes which
switched from add-on/retrofit devices to low-emission stoves showed decreases in
wood use. The average wood use for the add-on/retrofit vs. low-emission technology
decreased from 0.79 kg dry wood/HDD to 0.53 kg dry wood/HDD.
As noted in the evaluation of creosote accumulation, stove switching results are
intended to give qualitative results only. Nonetheless, the consistent reduction
of wood use by the advanced technology stoves indicates that wood use is reduced
with these stoves.
PARTICULATE EMISSIONS, BURN RATE, AND FUELING DATA
Introduction
One of the objectives of the original study design was to evaluate the emission
reduction performance of catalytic woodstoves, add-on/retrofit devices, and low-
emission stoves over a two-heating-season period. Tables 3-10A, 3-10B, and 3-10C
present data obtained for each sampling period in Group I and Group III homes
during the study. Data presented in Table 3-10A include stove codes, sampling
3-43
-------
Table 3-9
EFFECTS OF STOVE TECHNOLOGY CHANGES ON WOOD USEa/b/
CATALYTIC VS. CONVENTIONAL:
STUDY HOME
V19
V20
V22
V28
N18
N20
N22
Average
CONVENTIONAL STOVE
(Dry Kg Wood Use/HDD)
1.22
1.16
0.80
0.54
0.63
0.91
1.25
0.93
CATALYTIC STOVE
(Dry Kg Wood Use/HDD)
0.55
0.95
0.57
0.38
0.59
0.53
0.89
0.64
NET CHANGE IN
WOOD USE
-55
-18
-29
-30
-6
-42
-29
-30 [15]
ADD-ON/RETROFIT VS. CONVENTIONAL:
STUDY HOME
V15
V21
V29
N05
N27
Average
CONVENTIONAL STOVE
(Dry Kg Wood Use/HDD)
0.79
0.53
1.09
0.43
1.47
0.86
ADD-ON/RETROFIT
(Dry Kg Wood Use/HDD)
0.60
0.37
0.55
0.66
1.30
0.70
NET CHANGE IN
WOOD USE
-24
-30
-50
+53
-12
-13 [35]
LOW-EMISSION VS. CONVENTIONAL:
STUDY HOME
V23
N16
Average
CONVENTIONAL STOVE
(Dry Kg Wood Use/HDD)
0.51
0.60
0.56
LOW-EMISSION STOVE
(Dry Kg Wood Use/HDD)
0.36
0.39
0.38
NET CHANGE IN
WOOD USE
-29
-35
-32 [3]
(Continued)
3-44
-------
Table 3-9 (Continued)
EFFECTS OF STOVE TECHNOLOGY CHANGES ON WOOD USEa/b/
LOW-EMISSION VS. ADD-ON/RETROFIT:
STUDY HOME
V03
V12
N13
Average
ADD-ON/RETROFIT
(Dry Kg Wood Use/HDD)
0.62
0.53
1.23
0.79
LOW-EMISSION STOVE
(Dry Kg Wood Use/HDD)
0.59
0.60
0.39
0.53
NET CHANGE IN
WOOD USE
-5
+13
-68
-20 [35]
a/ Wood use data is from woodpile measurements only.
h/ Values inside brackets are standard deviations (o-).
3-45
-------
Figure 3-5
Comparative Wood Use: Group II Homes (Woodpile Measurements)
60 -
50 -
CDHPiiRATIUE
HDDD USE
REDUCTION
(X)
OJ
I
cr\
20 -
10 -
CftTflLVTIC STDUES
US .
CDMUENTIDNftL STQUES
(7 HOMES)
flDD-DN/RETRDFIIJ
US
CDMUENTIDNflL STDUES
(5 HOMES)
LDM EMISSION STQUES
US.
CDNUEHTIDNflL STDUES
(2 HOMES)
LDW EHISSIDH STDUES
US.
flDD-DN/RETRDFITS
(3 HOMES)
-------
Table 3-10A
STOVE USE CHARACTERISTICS — CATALYTIC STOVES
CO
-p»
Sampling
Code3/
V05(-l)t/
(-2)t/
_4ii/
_5u/
V07-1 ,
(_2)v/
(~5)w/
-6
-7
V08-1
(-3)v/
_4X/
_5x/
_6x/
_7x/
V11-2V/
-6Y/
-7V/
V13-2
_4U/x/
_5ii/x/
_6U/x/
-/u/x/
Stove
Codeb/
B
B
B
B
C
C
C
C
C
C
D
D
D
D
D
D
D
B
B
B
D
D
D
D
D
D
Sampling
Period0'
01/07-01/13/86
02/09-02/15/86
11/16-11/22/86
12/14-12/20/86
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
11/18-11/23/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
02/26-03/02/86
02/08-02/11/87
03/08-03/14/87
02/23-03/01/86
03/23-03/29/86
12/02-12/07/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
HDDd/
345
355
266
272
345
355
252
272
396
300
345
355
252
266
272
396
300
244
173
306
342
173
229
307
392
306
Catalyst
Operation6'
(%)
(96.7)
(96.3)
88.1
87.8
78.8
85.4
55.7
65.5
74.9
70.9
57.4
57.7
40.9
58.4
68.7
70.9
62.0
74.5
45.0
42.9
57.4
31.8
39.9
45.0
49.2
49.6
Stove
Operation^/
(%)
100.0
100.0
100.0
85.7
100.0
100.0
70.9
99.3
97.0
97.2
100.0
99.8
93.8
85.7
100.0
95.5
93.7
67.2
98.8
54.7
97.2
79.8
99.8
100.0
100.0
96.3
Heating
Systems/
Use (%)
3.1
1.7
0.0
2.8
25.0
39.1
38.3
15.9
37.6
11.2
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.4
0.0
0.9
Efficiency
(*)W
(46.9)
(50.0)
66.1
0.0
55.3
(51.0)
50.5
(54.9)
49.2
61.5
52.0
53.9
(41.3)
55.2
57.0
53.8
54.0
61.1
65.1
64.2
51.5
40.1
55.5
55.8
52.5
58.1
(Continued)
-------
Table 3-10A (Continued)
STOVE USE CHARACTERISTICS -- CATALYTIC STOVES
Cn
Sampl ing
Code3/
V16-1
-4U/z/
_5u/z/
-6"/z/
_7u/z/
V31-4Y/
V32-1Y/ ,
_5aa/y/
(N01-2)bb/
-3
-5
-6
-7
N02-1
(_2)cc/
_4X/
_6x/
_7x/
N03-4dd/
_i;dd/
_6u/dd/
N09-1 ,
-4U/
_6u/
_7u/
Stove
Code5/
C
C
C
C
C
P
P
P
A
A
A
A
A
D
D
D
D
D
D
C
C
C
B
B
B
B
Sampl ing
Periodc/
01/26-02/01/86
11/30-12/06/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
12/06-12/12/86
03/14-03/20/87
01/21-01/27/87
02/18-02/22/86
03/16-03/22/86
01/04-01/10/87
02/03-02/08/87
03/01-03/07/87
01/18-01/25/86
02/16-02/22/86
03/16-03/22/86
11/23-11/29/86
02/01-02/07/87
03/01-03/07/87
11/25-11/30/86
01/04-01/10/87
02/01-02/07/87
02/04-02/08/86
12/07-12/13/86
02/17-02/22/87
03/15-03/21/87
HDDd/
373
277
307
392
306
300
208
409
164
224
294
226
218
243
258
224
196
264
218
162
294
264
238
275
273
242
Catalyst
Operation6'
(%)
80.2
62.7
60.7
70.5
43.2
87.8
33.8
48.0
57.0
29.5
22.0
10.9
10.8
83.1
(50.5)
74.1
60.7
59.9
53.9
41.4
58.4
42.7
88.2
83.1
70.1
67.3
Stove
Operation' /
(%)
100.0
93.6
70.7
87.0
71.6
100.0
100.0
100.0
100.0
99.7
100.0
100.0
98.8
100.0
100.0
98.2
98.2
100.0
100.0
40.6
76.5
66.4
99.7
100.0
99.0
84.8
Heating
System?/
Use (%)
0.0
6.0
2.7
23.4
9.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.2
1.0
1.3
0.0
0.4
0.0
0.5
Efficiency
(%)W
58.7
48.2
48.7
52.0
50.6
48.2
51.4
62.5
(71.8)
55.1
48.4
46.3
50.8
59.0
(44.8)
59.9
59.6
57.9
57.3
55.2
52.3
44.9
58.3
48.7
52.2
56.7
(Continued)
-------
Table 3-10A (Continued)
STOVE USE CHARACTERISTICS — CATALYTIC STOVES
-P.
ID
Sampling
Code3/
Nio-iy/ ,
(_2)v/y/
(_4)V/Y/
-5Y/
-&y/
-7y/
N11-2V/ ,
_4*/y/
(_6)v/x/y/
(-7)v/x/y/
N18-4u/y/
_5u/y/
_6u/y/
-?y/
(N32-3)v/y/
-sy/
N33-3 ,
_5ee/
Stove
Codeb/
A
A
A
A
A
A
D
D
D
D
B
B
B
B
P
P
P
P
Sampling
Period0'
02/02-02/08/86
03/02-03/08/86
12/09-12/14/86
01/18-01/24/87
02/15-02/21/87
03/15-03/21/87
03/02-03/08/86
12/07-12/13/86
02/17-02/22/86
03/15-03/21/86
11/23-11/29/86
01/04-01/10/87
02/03-02/08/87
03/01-03/07/87
03/02-03/08/87
01/06-01/11/87
02/27-03/05/86
01/18-01/24/87
HDDd/
249
263
241
331
376
242
263
275
273
242
196
294
226
218
263
264
255
331
Catalyst
Operation6'
(%)
91.1
80.3
82.3
90.3
86.6
87.2
31.7
53.7
65.0
36.2
85.1
85.0
89.6
90.9
76.8
51.3
58.8
(0.0)
Stove
Operation''
(%)
100.0
98.5
100.0
100.0
100.0
100.0
88.9
93.3
100.0
52.1
100.0
100.0
78.6
98.1
100.0
100.0
100.0
100.0
Heating
SystemS/
Use (%)
2.8
0.1
0.0
0.0
0.0
0.0
21.9
10.0
8.5
15.3
0.0
0.4
0.0
0.0
0.0
0.0
0.0
0.0
Efficiency
(%)"/
67.7
(58.7)
(63.0)
58.0
60.6
50.5
56.6
62.1
(73.5)
(54.0)
60.4
49.4
50.8
52.9
(61.9)
56.1
50.0
49.8
(Continued)
-------
Table 3-10A (Continued)
STOVE USE CHARACTERISTICS -- ADD-ON/RETROFIT STOVES
en
O
Samp] ing
Codea/
V01-4"/
-5U/
(-6)u/ff/
-7U/
V02-1
_2
V03-1
-3
V1°-2 ,uu/
-sgg/hh/
-699/hh/
(Vl2-l)v/ee/
_2ee/
-3
(V15_l)w/y/
N04-lii/
-511/
(_6)v/ii/
N06-1
-2
-3
N12-4
N14-2
(_4)u/v/jj/
(_5)u/v/jj/
Stove
Codeb/
-^IRT
E'(R)
E (R)
E (R)
G (A)
G (A)
F (R)
F (R)
H (A)
J (A)
J (A)
F (R)
F (R)
F (R)
H (A)
G (A)
J (A)
G (A)
I (A)
I (A)
I (A)
J (A)
I (A)
J (A)
0 (A)
Sampl ing
Period0'
11/16-11/22/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
01/07-01/13/86
02/13-02/17/87
01/07-01/13/86
03/09-03/15/86
02/23-03/01/86
01/11-01/17/87
02/08-02/14/87
01/26-02/01/86
02/20-02/26/86
03/23-03/29/86
01/28-02/01/86
01/19-01/25/86
01/04-01/10/87
02/01-02/07/87
01/19-01/25/86
02/16-02/22/86
03/16-03/22/86
12/07-12/13/86
03/02-03/08/86
12/07-12/13/86
01/18-01/24/87
HDDd/
266
272
396
300
345
249
345
252
342
307
392
373
316
173
292
243
294
264
243
258
224
275
263
275
331
Catalyst
Operation6'
(%)
69.2
74.7
(75.3)
63.9
37.5
49.5
25.5
17.7
19.0
17.6
56.2
(0.0)
(0.0)
9.8
22.4
46.4
57.8
9.4
53.2
66.6
56.3
53.6
53.7
68.2
N/A
Stove
Operation"'"/
(%)
99.9
100.0
100.0
95.4
100.0
99.2
99.9
98.5
89.1
74.6
73.8
99.6
98.2
47.9
98.5
92.4
96.1
99.3
99.3
100.0
84.7
93.9
99.9
100.0
100.0
Heating
System?/
Use (%)
0.3
1.9
0.0
2.3
0.0
0.0
0.0
2.6
5.5
4.7
5.3
0.0
0.0
0.0
0.0
2.7
0.1
0.1
0.0
0.0
0.0
0.2
0.0
0.0
0.0
Efficiency
(%)W
62.2
58.0
(54.6)
60.2
56.4
58.8
53.3
38.3
50.7
54.5
73.8
(58.5)
51.5
32.8
(58.1)
54.5
61.4
(59.6)
52.3
55.5
45.6
59.3
49.5
(63.2)
(57.0)
(Continued)
-------
Table 3-10A (Continued)
STOVE USE CHARACTERISTICS — LOW-EMISSION STOVES
CO
I
un
Sampling
Code3/
V03-5
-6
(V04-l)v/
-3
-4
-6
V12-6U/
V14-6
-7
V18-4y/
-sy/
x
-?y/
V34-5U/
_7u/
V35-?y/
N07-5 ,
(-6)v/
-7
(N13-5)v/y/
N15.4U/kk/
_5ii/kk/
_7u/kk/
N16-4y/
-&l
-?y/
Stove
Code5/
N
N
L
L
L
L
M
M
M
K
K
K
K
M
M
N
K
K
K
M
L
L
L
N
N
N
Sampling
Period0'
12/14-12/20/86
01/25-01/31/87
01/07-01/13/86
03/09-03/15/86
11/16-11/22/86
01/25-01/31/87
02/08-02/14/87
02/08-02/14/87
03/08-03/14/87
11/30-12/06/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
12/14-12/20/87
02/22-02/28/87
03/08-03/14/87
01/04-01/10/87
02/01-02/07/87
03/01-03/07/87
01/18-01/24/87
12/07-12/13/87
01/18-01/24/87
03/15-03/21/87
12/07-12/13/86
02/15-02/21/87
03/15-03/21/87
HDDd/
272
396
345
252
376
396
392
392
306
277
307
392
306
272
300
306
294
264
218
331
275
331
242
275
376
242
Stove
Operation'/
(%)
100.0
99.5
95.1
80.3
66.9
91.3
100.0
99.3
100.0
82.8
96.0
87.5
97.5
100.0
100.0
84.1
100.0
97.3
100.0
100.0
88.1
98.5
84.7
97.4
100.0
100.0
Heating
System^/
Use (%)
1.8
2.5
0.0
2.3
2.0
0.0
0.0
0.0
0.0
1.2
0.1
7.7
0.8
1.0
1.9
0.0
0.0
0.0
0.0
0.0
0.5
11.1
4.0
0.0
0.0
0.1
Efficiency
(*)W
48.0
47.8
(52.3)
48.2
38.5
42.4
55.5
46.0
51.0
44.5
49.2
50.8
39.8
54.3
56.9
52.6
57.3
(40.5)
59.0
45.4
33.0
48.4
45.9
48.9
66.5
48.2
(Continued)
-------
Table 3-10A (Continued)
STOVE USE CHARACTERISTICS -- TRADITIONAL/CONVENTIONAL STOVES
CO
I
IV)
Sampl ing
Code3/
V06-1V/
-2Y/
(_3)V/y/
-5Y/
-ey/
V09-1
V14-1
-2
-3
(N08-3)v/
_4U/
_6u/
_7u/
N14-6
-7
N16-1V/
Stove
Codeb/
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Sampl ing
Period0'
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
12/14-12/20/86
01/25-01/31/87
01/26-02/01/86
01/26-02/01/86
02/26-03/02/86
03/23-03/29/86
03/16-03/22/86
11/25-11/30/86
02/01-02/07/87
03/01-03/07/87
02/15-02/21/87
03/15-03/21/87
02/02-02/08/87
HDDd/
345
355
252
272
396
373
373
244
173
189
162
264
218
376
242
321
Stove
Operation"*/
(%)
95.6
100.0
93.9
99.8
100.0
95.9
100.0
100.0
77.6
100.0
100.0
100.0
100.0
100.0
100.0
100.0
Heating
SystemS/
Use (%)
0.0
0.0
0.0
0.0
0.0
0.0
3.3
3.1
0.0
0.1
0.0
0.0
0.0
0.0
0.0
3.8
Efficiency
(%)W
53.5
(50.2)
41.5
43.7
30.6
52.6
54.6
48.7
48.6
(53.5)
49.8
53.8
51.8
53.9
52.6
49.5
-------
Table 3-10B
FUEL CHARACTERISTICS — CATALYTIC STOVES
OJ
I
tn
uo
Sampling
Code3/
V05(-l)t/
(-2)*/
_4U/
_5u/
V07-1 ,
(_2)v/
(~5)w/
-6
-7
V08-1
(:3)V/
_4X/
_5x/
_6x/
_7x/
V11-2V/
-6Y/
-7V/
V13-2
_3
_4U/x/
_5U/x/
_6u/x/
_yu/x/
Stove
Codeb/
B
B
B
B
C
C
C
C
C
C
D
D
D
D
D
D
D
B
B
B
D
D
D
D
D
D
Sampling
Period0/
01/07-01/13/86
02/09-02/15/86
11/16-11/22/86
12/14-12/20/86
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
11/18-11/23/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
02/26-03/02/86
02/08-02/11/87
03/08-03/14/87
02/23-03/01/86
03/23-03/29/86
12/02-12/07/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
HDDd/
345
355
266
272
345
355
252
272
396
300
345
355
252
266
272
396
300
244
173
306
342
173
229
307
392
306
Fuel
Moisture1'
(% DB)
28.0
29.5
18.2
19.0
17.0
20.5
25.0
20.1
12.5
13.8
33.0
25.0
28.5
33.0
26.0
28.4
29.0
27.0
28.0
24.0
25.0
24.5
23.0
20.5
19.8
27.0
Average
LoadJ/
(kg dry)
(7.4)
(8.1)
4.4
6.0
8.4
10.2
9.5
(7.4)
8.5
9.1
5.7
6.1
4.9
3.8
4.3
5.2
5.2
14.0
9.0
15.1
4.4
3.2
4.1
4.3
4.1
3.9
Loading
Frequency*/
(#/hr)
(0.18)
(0.17)
0.19
0.15
0.22
0.15
0.11
(0.10)
0.16
0.16
0.21
0.21
0.20
0.16
0.18
0.21
0.21
0.07
0.13
0.08
0.25
0.22
0.29
0.27
0.30
0.30
Burn
Rate1/
(kg/hr)
(1.33)
(1.19)
0.84
0.92
1.85
1.89
1.47
(0.76)
1.35
1.45
1.22
1.27
0.96
0.61
0.79
1.08
1.09
1.02
1.19
1.15
1.10
0.73
1.21
1.18
1.24
1.18
(Continued)
-------
Table 3-10B (Continued)
FUEL CHARACTERISTICS -- CATALYTIC STOVES
Sampl ing
Code9/
V16-1
_4U/z/
_5u/z/
-6U/Z/
-/u/z/
V31-4W
V32-1Y/
-5aa/y/
(N01-2)bb/
-3
-5
-6
7
N02-1
(_2)cc/
_4X/
_6x/
_7x/
N03-4dd/
_5dd/
_6u/dd/
N09-1 ,
_4U/
_6u/
_7u/
Stove
Code5/
C
C
C
C
C
P
P
P
A
A
A
A
A
D
D
D
D
D
D
C
C
C
B
B
B
B
Sampl ing
Period0/
01/26-02/01/86
11/30-12/06/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
12/06-12/12/86
03/14-03/20/87
01/21-01/27/87
02/18-02/22/86
03/16-03/22/86
01/04-01/10/87
02/03-02/08/87
03/01-03/07/87
01/18-01/25/86
02/16-02/22/86
03/16-03/22/86
11/23-11/29/86
02/01-02/07/87
03/01-03/07/87
11/25-11/30/86
01/04-01/10/87
02/01-02/07/87
02/04-02/08/86
12/07-12/13/86
02/17-02/22/87
03/15-03/21/87
HDDd/
373
277
307
392
306
300
208
409
164
224
294
226
218
243
258
224
196
264
218
162
294
264
238
275
273
242
Fuel
Moisture1/
(% DB)
18.5
29.2
26.6
27.2
30.3
32.4
24.0
26.2
31.6
29.0
38.8
43.0
39.3
15.6
16.7
16.7
14.6
11.8
15.0
16.7
16.7
20.2
41.0
15.8
17.1
16.7
Average
LoadJ/
(kg dry)
7.5
6.5
5.5
6.2
6.2
2.9
4.1
6.4
4.3
4.5
4.5
4.9
4.4
5.3
(7.1)
5.6
5.7
5.8
4.8
5.3
5.7
5.5
10.5
11.6
8.8
11.5
Loading
Frequency1"-/
(#/hr)
0.17
0.17
0.23
0.20
0.18
0.35
0.27
0.18
0.16
0.13
0.17
0.15
0.16
0.22
(0.21)
0.19
0.15
0.20
0.20
0.19
0.17
0.19
0.12
0.11
0.14
0.12
Burn
Rate1/
(kg/hr)
1.29
1.11
1.25
1.28
1.13
1.01
1.12
1.14
0.68
0.57
0.78
0.74
0.72
1.17
(1-49)
1.05
0.82
1.17
0.97
1.00
0.97
1.03
1.23
1.31
1.23
1.37
(Continued)
-------
Table 3-10B (Continued)
FUEL CHARACTERISTICS -- CATALYTIC STOVES
oo
r
Sampling
Code3/
N10-iy/ ,
(-2)V/y/
(_4)v/y/
-5y/
-&l
-1^1
Nll-2y/ ,
-4x/y/
(_6)v/*/y/
(_7)v/x/y/
N18-4u/y/
_5u/y/
-eu/y/
-?y/
(N32-3)V/y/
-sy/
N33-3 ,
_5ee/
Stove
Code5/
A
A
A
A
A
A
D
D
0
D
B
B
B
B
P
P
P
P
Sampling
Period0'
02/02-02/08/86
03/02-03/08/86
12/09-12/14/86
01/18-01/24/87
02/15-02/21/87
03/15-03/21/87
03/02-03/08/86
12/07-12/13/86
02/17-02/22/86
03/15-03/21/86
11/23-11/29/86
01/04-01/10/87
02/03-02/08/87
03/01-03/07/87
03/02-03/08/87
01/06-01/11/87
02/27-03/05/86
01/18-01/24/87
HDDd/
249
263
241
331
376
242
263
275
273
242
196
294
226
218
263
264
255
331
Fuel
Moisture1'
(% DB)
36.0
41.4
26.0
39.4
37.2
37.0
17.8
16.6
15.7
16.5
11.0
16.6
16.6
17.8
23.5
22.0
32.3
30.0
Average
LoadJ/
(kg dry)
5.3
4.7
6.9
7.2
8.1
11.1
2.9
2.4
2.9
2.7
7.6
9.1
5.9
7.4
6.5
7.4
6.8
6.9
Loading
Frequency14'
(#/hr)
0.28
0.25
0.17
0.21
0.21
0.14
0.31
0.38
0.40
0.22
0.18
0.17
0.20
0.19
0.17
0.16
0.27
0.33
Burn
Rate1/
(kg/hr)
1.46
1.16
1.16
1.51
1.69
1.58
0.90
0.90
1.16
0.58
1.37
1.57
1.19
1.40
1.09
1.18
1.83
2.26
(Continued)
-------
Table 3-10B (Continued)
FUEL CHARACTERISTICS -- ADD-ON/RETROFIT STOVES
en
cr>
Sampl ing
Code3/
V01-4U/
-5U/
(-6)"/ff/
_7u/
V02-1
-2
V03-1
-3
-599/hh/
-699/hh/
(V12-l)v/
-2
(V15-l)w'>/
N04-1! !/,
-511/
(_6)v/ii/
N06-1
-2
-3
N12-4
N14-2
(_4)U/v/jj/
.5 u/v/jj/
Stove
Codeb/
E (R)
E (R)
E (R)
E (R)
G (A)
G (A)
F (R)
F (R)
H (A)
J (A)
J (A)
F (R)
F (R)
F (R)
H (A)
G (A)
J (A)
G (A)
I (A)
I (A)
I (A)
J (A)
I (A)
J (A)
0 (A)
Sampl ing
Period0'
11/16-11/22/86
12/14-12/20/86
01/25-01/31/87
02/22-02/28/87
01/07-01/13/86
02/13-02/17/87
01/07-01/13/86
03/09-03/15/86
02/23-03/01/86
01/11-01/17/87
02/08-02/14/87
01/26-02/01/86
02/20-02/26/86
03/23-03/29/86
01/28-02/01/86
01/19-01/25/86
01/04-01/10/87
02/01-02/07/87
01/19-01/25/86
02/16-02/22/86
03/16-03/22/86
12/07-12/13/86
03/02-03/08/86
12/07-12/13/86
01/18-01/24/87
HDDd/
266
272
396
300
345
249
345
252
342
307
392
373
316
173
292
243
294
264
243
258
224
275
263
275
331
Fuel
Moisture1'
(% DB)
17.5
27.1
34.0
34.0
32.0
28.0
24.5
24.7
21.7
24.0
22.0
21.0
31.0
21.0
34.0
14.2
19.1
16.3
21.7
22.3
22.5
27.6
42.0
25.4
30.0
Average
LoadJ/
(kg dry)
5.3
6.4
6.2
6.4
10.5
12.0
7.6
5.3
4.0
3.8
4.0
4.9
3.8
3.9
(5.2)
8.2
7.9
7.9
7.3
9.2
8.0
6.9
6.9
6.0
5.4
Loading
Frequencyk/
(#/hr)
0.22
0.21
0.26
0.21
0.15
0.13
0.21
0.16
0.23
0.28
0.37
0.28
0.25
0.12
(0.06)
0.21
0.21
0.23
0.32
0.25
0.26
0.19
0.34
0.30
0.40
Burn
Rate1/
(kg/hr)
1.17
1.36
1.58
1.37
1.62
1.61
1.59
0.87
1.01
1.07
1.47
1.37
0.97
0.97
(0.29)
1.70
1.66
1.86
2.32
2.29
2.08
1.31
2.35
1.78
2.16
(Continued)
-------
Table 3-10B (Continued)
FUEL CHARACTERISTICS — LOW-EMISSION STOVES
Sampling
Code3/
V03-5
(V04-l)v/
-3
-4
V12-6U/
V14-6
-7
V18-4J"
-5Y/
-ey/
-7y/
V34-5U/,
-7U/
V35-7*/
N07-5 ,
(-6)v/
-7 ,
(N13-5)V,/W
N15_4u/kk/
_5U/kk/
_7u/kk/
N16-4y/,
-ey/
-7V/
Stove,
Codeb/
N
N
L
L
L
L
M
M
M
K
K
K
K
M
M
N
K
K
K
M
L
L
L
N
N
N
Sampl ing
Period0'
12/14-12/20/86
01/25-01/31/87
01/07-01/13/86
03/09-03/15/86
11/16-11/22/86
01/25-01/31/87
02/08-02/14/87
02/08-02/14/87
03/08-03/14/87
11/30-12/06/86
01/11-01/17/87
02/08-02/14/87
03/08-03/14/87
12/14-12/20/87
02/22-02/28/87
03/08-03/14/87
01/04-01/10/87
02/01-02/07/87
03/01-03/07/87
01/18-01/24/87
12/07-12/13/87
01/18-01/24/87
03/15-03/21/87
12/07-12/13/86
02/15-02/21/87
03/15-03/21/87
HDDd/
272
396
345
252
376
396
392
392
306
277
307
392
306
272
300
306
294
264
218
331
275
331
242
275
376
242
Fuel
Moisture1/
(% DB)
34.0
33.0
21.0
15.2
15.0
13.0
27.0
28.3
27.0
31.0
34.3
34.0
26.0
21.1
21.0
35.5
20.2
21.1
20.9
32.5
15.5
15.5
15.5
25.0
23.2
24.0
Average
LoadJ/
(kg dry)
5.2
4.6
2.8
2.8
2.5
2.5
2.2
3.7
3.5
3.1
3.8
3.5
4.0
2.8
3.8
3.1
5.7
4.6
4.6
3.9
2.2
2.7
2.9
2.9
3.3
3.2
Loading
Frequency*/
(#/hr)
0.24
0.30
0.38
0.33
0.31
0.32
0.30
0.29
0.24
0.35
0.29
0.31
0.32
0.27
0.24
0.29
0.15
0.18
0.20
0.31
0.53
0.50
0.32
0.34
0.33
0.27
Burn
Rate1/
(kg/hr)
1.28
1.38
1.07
0.90
0.76
0.81
0.67
1.07
0.85
1.09
1.10
1.08
1.26
0.76
0.92
0.90
0.84
0.85
0.90
1.18
1.19
1.34
0.93
0.97
1.10
0.87
CO
I
U1
•—I
(Continued)
-------
Table 3-10B (Continued)
FUEL CHARACTERISTICS -- TRADITIONAL/CONVENTIONAL STOVES
_
Co
Sampl ing
Codea/
V06-1V/
-2V/
(_3)v/y/
-5Y/
-6Y/
V09-1
V14-1
-2
-3
(N08-3)V/
_4U/
_5u/
_7u/
N14-6
-7
N16-1Y/
Stove
Codeb/
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Sampl ing
Period0/
01/07-01/13/86
02/09-02/15/86
03/09-03/15/86
12/14-12/20/86
01/25-01/31/87
01/26-02/01/86
01/26-02/01/86
02/26-03/02/86
03/23-03/29/86
03/16-03/22/86
11/25-11/30/86
02/01-02/07/87
03/01-03/07/87
02/15-02/21/87
03/15-03/21/87
02/02-02/08/87
HDDd/
345
355
252
272
396
373
373
244
173
189
162
264
218
376
242
321
Fuel
Moisture1'
(% DB)
26.5
26.6
27.5
25.0
28.0
41.2
23.0
28.2
26.3
29.7
24.7
26.5
29.5
35.2
41.0
26.0
Average
loadJ/
(kg dry)
8.0
6.7
7.4
6.7
7.3
3.8
4.7
4.7
3.5
9.8
7.2
7.8
8.4
5.6
5.4
3.5
Loading
Frequency^/
(#/hr)
0.31
0.24
0.22
0.23
0.26
0.30
0.35
0.31
0.26
0.20
0.26
0.28
0.24
0.43
0.29
0.45
Burn
Rate1/
(kg/hr)
2.45
1.60
1.59
1.52
1.86
1.12
1.67
1.45
0.92
1.92
1.91
2.19
2.00
2.45
1.57
1.55
-------
Table 3-10C
EMISSION CHARACTERISTICS -- CATALYTIC STOVES
Sampling
Code3/
V05(-l)t/
(-2)*/
_4U/
_5u/
V07-1 ,
(_2)v/
(-5)w/
-6
-7
V08-1
(:$'
_5x/
-6X/
_7x/
V11-2V/
-ey/
-7y/
V13-2
_4U/x/
_5u/x/
_6"/x/
_7u/x/
Stove,
Codeb/
B
B
B
B
C
C
C
C
C
C
D
D
D
D
D
D
D
B
B
B
D
D
D
D
D
D
HDDd/
345
355
266
272
345
355
252
272
396
300
345
355
252
266
272
396
300
244
173
306
342
173
229
307
392
306
Burn
Rate1/
(kg/hr)
(1.33)
(1.19)
0.84
0.92
1.85
1.89
1.47
(0.76)
1.35
1.45
1.22
1.27
0.96
0.61
0.79
1.08
1.09
1.02
1.19
1.15
1.10
0.73
1.21
1.18
1.24
1.18
Particulate Emissions111'
(g/hr)n/
(33.7)
(18.5)
9.0
31.4
10.1
(10.3)
11.4
(4.8)
14.3
1.7
18.2
20.4
(25.4)
7.6
14.1
13.4
12.7
6.1
6.3
7.0
17.8
19.4
12.4
9.7
12.4
10.5
(g/kg)°/
(25.4)
(15.6)
10.8
34.1
5.5
(5.4)
7.7
(6.3)
10.5
1.2
15.0
16.1
(26.3)
12.4
17.8
12.4
11.7
6.0
5.3
6.1
16.1
26.7
10.3
8.1
10.0
8.9
(g/106J)P/
(2.7)
(1-6)
0.9
0.0
0.5
(0.5)
0.8
(0.6)
1.1
0.1
1.4
1.5
(3.2)
1.1
1.6
1.2
1.1
0.5
0.4
0.5
1.6
3.3
1.0
0.7
1.0
0.8
(g/m3)q/
(0.73)
(0.58)
0.45
1.02
0.29
0.24
0.29
0.26
0.43
0.06
1.04
1.22
1.13
0.68
1.31
0.96
0.78
0.24
0.30
0.29
0.85
0.88
0.61
0.44
0.52
0.53
Average
Flue 0?r/
(%)
(17.9)
(17.1)
16.5
17.8
15.4
(16.2)
16.9
16.6
16.6
15.7
13.7
13.1
(16.5)
15.2
13.3
12.9
14.0
16.8
15.0
16.0
15.5
17.5
14.7
15.3
15.6
14.7
Average
Flue Temp5'
(°C)
(103.3)
(148.5)
70.7
61.6
231.5
232.3
189.8
179.9
204.2
214.6
225.7
211.6
190.6
171.9
170.0
249.2
225.1
131.6
144.8
126.2
180.9
160.8
205.3
206.1
213.5
187.6
(Continued)
-------
Table 3-10C (Continued)
EMISSION CHARACTERISTICS -- CATALYTIC STOVES
Sampl ing
Code9/
V16-1
_4U/z/
_5u/z/
_6u/z/
_7U/z/
V31-4V/
V32-lW ,
_5aa/y/
(N01-2)bb/
-3
-5
-6
-7
N02-1
(_2)cc/
_4X/
_6x/
_7x/
N03-4dd/
-5dd/
_6u/dd/
N09-1 ,
_4U/
-6"/
_7u/
Stove
Codeb/
C
C
C
C
C
P
P
P
A
A
A
A
A
D
D
D
D
D
D
C
C
C
B
B
B
B
HDDd/
373
277
307
392
306
300
208
409
164
224
294
226
218
243
258
224
196
264
218
162
294
264
238
275
273
242
Burn
Rate1/
(kg/hr)
1.29
1.11
1.25
1.28
1.13
1.01
1.12
1.14
0.68
0.57
0.78
0.74
0.72
1.17
(1.49)
1.05
0.82
1.17
0.97
1.00
0.97
1.03
1.23
1.31
1.23
1.37
Particulate Emissions'11/
(g/hr)n/
8.2
19.0
21.8
16.8
15.1
17.7
13.9
11.8
(4.6)
13.0
21.9
21.2
15.9
9.9
(44.5)
7.0
6.9
10.0
8.4
8.1
19.0
24.3
15.7
21.2
17.1
29.6
(g/kg)°/
6.3
17.1
17.4
13.1
13.3
17.5
12.4
10.3
(6.8)
22.9
28.1
28.6
22.2
8.5
(29.9)
6.7
8.4
8.6
8.7
8.1
19.7
23.6
12.8
16.2
13.9
21.5
(g/106J)P/
0.5
1.8
1.8
1.3
1.3
1.8
1.2
0.8
(0.5)
2.1
2.9
3.1
2.2
0.7
(3.3)
0.6
0.7
0.7
0.7
0.7
1.9
2.6
1.1
1.6
1.4
1.9
(g/m3)q/
0.36
0.83
0.82
0.70
0.63
0.76
0.54
0.71
0.79
1.10
1.41
1.07
0.98
0.71
(2.44)
0.44
0.61
0.66
0.63
0.28
0.91
0.74
0.86
1.08
0.89
1.41
Average
Flue 02r/
(%)
15.0
15.9
16.0
15.4
16.0
16.4
16.4
13.8
(8.8)
16.0
15.7
17.1
16.3
12.2
(12.3)
14.1
13.3
12.9
13.3
17.3
16.1
17.7
14.0
14.0
14.3
14.1
Average
rlue Temp5/
(°c)
203.6
192.9
184.8
197.7
192.2
171.1
177.5
141.4
96.4
84.5
120.6
110.6
119.6
257.4
(252.1)
215.6
221.3
256.5
245.9
146.2
139.8
133.8
152.6
268.3
232.1
110.0
(Continued)
-------
Table 3-IOC (Continued)
EMISSION CHARACTERISTICS — CATALYTIC STOVES
Sampling
Code3/
N10-iy/ ,
(_2)v/y/
(_4)v/y/
-&l
-&l
-?y/
M11:$y/
(_6)v/x/y/
(_7)v/x/y/
N18-4u/y/
_5u/y/
_6u/y/
-7V/
(N32-3)v/y/
-5y/
N33-3 ,
-5ee/
Stove
Codeb/
A
A
A
A
A
A
D
D
D
D
B
B
B
B
P
P
P
P
HDDd/
249
263
241
331
376
242
263
275
273
242
196
294
226
218
263
264
255
331
Burn
Rate '/
(kg/hr)
1.46
1.16
1.16
1.51
1.69
1.58
0.90
0.90
1.16
0.58
1.37
1.57
1.19
1.40
1.09
1.18
1.83
2.26
Particulate Emissions111'
(g/hr)n/
9.7
(17.9)
(13.9)
23.4
18.2
39.7
14.9
5.5
(6.6)
(4.6)
20.6
41.3
31.6
29.2
(5.4)
19.6
22.3
34.6
(g/kg)°/
6.7
(15.4)
(12.0)
15.5
10.8
25.1
16.7
6.0
(5.7)
(7.9)
15.1
26.4
26.5
20.8
(5.0)
16.6
12.2
15.3
(g/106J)P/
0.5
(1.3)
(0.9)
1.3
0.9
2.5
1.4
0.5
(0.4)
(0.7)
1.2
2.6
2.5
1.9
(0.4)
1.5
1.2
1.5
(g/m3)^
0.62
1.17
1.26
1.55
0.90
1.86
1.40
0.43
0.48
0.37
1.24
1.64
1.92
1.64
0.29
0.97
0.57
0.85
Average
Flue 02r/
(%)
11.2
(13.1)
(10.1)
10.6
12.3
13.2
12.1
13.4
(12.1)
(16.1)
12.0
14.3
13.1
12.5
(14.8)
14.9
16.1
15.2
Average
Flue Temp5/
(°C)
143.6
134.5
161.3
181.7
175.4
161.7
189.2
212.3
89.5
200.4
159.9
165.3
164.5
195.4
186.1
143.9
197.5
208.1
CO
I
(Continued)
-------
Table 3-10C (Continued)
EMISSION CHARACTERISTICS -- ADD-ON/RETROFIT STOVES
Sampl ing
Code3/
V01-4U/
_5u/
(_6)u/ff/
_7u/
V02-1
-2
V03-1
-3
V10-2
_5gg/hh/
-699/hh/
(V12-l)v/
-2
-3
(V15-1)W/V/
N04-111/
.51 i/
(_6)v/ii/
N06-1
-2
-3
N12-4
N14-2
(_4)u/v/jj/
(_5)u/v/jj/
Stove
Codeb/
E (R)
E (R)
E (R)
E (R)
G (A)
G (A)
F (R)
F (R)
H (A)
J (A)
J (A)
F (R)
F (R)
F (R)
H (A)
G (A)
J (A)
G (A)
I (A)
I (A)
I (A)
J (A)
I (A)
J (A)
0 (A)
HDDd/
266
272
396
300
345
249
345
252
342
307
392
373
316
173
292
243
294
264
243
258
224
275
263
275
331
Burn
Rate1/
(kg/hr)
1.17
1.36
1.58
1.37
1.62
1.61
1.59
0.87
1.01
1.07
1.47
1.37
0.97
0.97
(0.29)
1.70
1.66
1.86
2.32
2.29
2.08
1.31
2.35
1.78
2.16
Particulate Emissions"1/
(g/hr)"/
6.3
10.1
(16.7)
7.1
17.1
15.5
18.6
31.8
16.2
21.3
8.4
(16.1)
16.5
36.7
(2.1)
18.7
14.2
(13.9)
16.9
13.6
37.3
7.3
25.7
(11.9)
(27.3)
(g/kg)°/
5.3
7.5
(10.6)
5.3
10.5
9.6
11.7
36.5
16.1
19.9
5.7
(11.8)
17.1
37.9
(7.1)
11.0
8.6
(7.5)
7.3
5.9
17.9
5.5
10.9
(6.7)
(12.7)
(g/106J)P/
0.4
0.7
(i.o)
0.4
0.9
0.8
1.1
4.7
1.6
1.8
0.4
(i.o)
1.7
5.8
(0.6)
1.0
0.7
(0.6)
0.7
0.5
1.9
0.5
1.1
(0.5)
(1.1)
(g/m3)q/
0.35
0.52
(0.74)
0.36
0.69
0.72
0.53
1.06
0.62
1.26
0.41
0.92
0.87
1.31
0.35
0.76
0.79
0.42
0.47
0.41
0.88
0.33
0.56
0.64
1.14
Average
Flue 02r/
(%)
14.3
13.9
14.0
14.1
14.3
13.5
16.4
18.0
17.0
14.5
13.7
(13.1)
15.8
17.4
16.0
13.9
11.6
(15.1)
14.4
13.9
15.9
14.8
15.6
(11.1)
(11.8)
Average
Flue Temp5/
(°c)
200.0
226.2
227.1
212.4
196.8
199.3
166.4
123.8
145.4
138.7
67.3
196.1
162.1
183.2
164.6
242.6
226.1
181.8
280.0
273.1
216.8
202.1
217.6
227.9
222.5
(Continued)
-------
Table 3-10C (Continued)
EMISSION CHARACTERISTICS -- LOW-EMISSION STOVES
Sampling
Code3/
V03-5
(V04-l)v/
:!
V12-6U/
V14-6
V18-4W
:$#
V34-5^
V35-71"
»<^,v/
(N13-5)VW
«»:i5ffi
_7ii/kk/
N16-4^
i?y/
Stove
Codeb/
N
N
L
L
L
M
M
M
K
K
K
K
M
M
N
I
K
M
L
L
N
N
N
HDDd/
272
396
376
396
392
We
277
307
392
306
272
300
306
218
331
Wl
242
275
376
242
Burn
Rate1/
(kg/hr)
1.28
1.38
1.07
0.90
0-76
U .01
0.67
Dig?
1.09
1.10
1.08
1.26
0.76
0.92
0.90
Sit
0.90
1.18
1.19
0!93
0.97
1.10
0.87
Particulate Emissions1"/
(g/hr)n/
18.3
2.0
(l:l]
14.1
6.5
5.2
26.3
17.2
17i3
47.6
7.9
5.9
3.6
12.9
25.2
9.4
(13.3)
9.4
ll!4
10.0
4.3
10.3
(9/kg)°/
14.3
1.4
2.4
0.9
0.7
2.7
2.0
2.9
2.3
0.9
0.6
0.4
(o§
(1.2)
1.2
0.6
1.3
J:§
1.2
(g/m3)q/
0.63
0.05
0.39
0.34
0.64
0.28
0.38
1.04
0.94
Oi94
2.18
0!30
0.15
1.18
1.17
0.71
0.50
0.25
0.32
0.51
0.47
0.40
0.45
Average
Flue 0?r/
(%)
16.6
17.2
(15.2)
B:S
16.1
16.8
16.4
15i2
15.3
15.7
16.2
17.1
,13.3,
17.0
14.2
16.6
&i
16.7
16.4
10.7
17.2
Average
Flue Temp5/
(°C)
181.1
222.8
252.8
230.9
203.1
215.4
182.4
146.1
132.9
19l!4
179.8
188.1
182.5
183.2
170.8
168.9
180.7
219.3
251.8
286.9
214.2
209.1
234.6
175.8
(Continued)
-------
Table 3-10C (Continued)
EMISSION CHARACTERISTICS -- TRADITIONAL/CONVENTIONAL STOVES
Sampl ing
Codea/
V06-1Y/
-2Y/
(-3)V/Y/
-5Y
-6V/
V09-1
V14-1
-2
-3
(N08-3)v/
_4U/
-6«/
_7u/
N14-6
-7
N16-1V/
Stove
Codeb/
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HDDd/
345
355
252
272
396
373
373
244
173
189
162
264
218
376
242
321
Burn
Rate1/
(kg/hr)
2.45
1.60
1.59
1.52
1.86
1.12
1.67
1.45
0.92
1.92
1.91
2.19
2.00
2.45
1.57
1.55
Particulate Emissions'11/
(g/hr)"/
2.9
4.7
(30.4)
12.7
17.3
15.4
16.9
23.5
20.3
(26.5)
32.6
26.6
30.9
34.0
29.0
13.9
(g/kg)°/
1.2
2.9
(19.1)
8.4
9.3
13.7
10.2
16.3
22.0
(13.8)
17.1
12.2
15.4
13.9
18.4
9.0
(g/106J)P/
0.1
0.3
(2.3)
1.0
1.5
1.3
0.9
1.7
2.3
(1.3)
1.7
1.2
1.5
1.3
1.7
0.9
(g/n)3)q/
0.06
0.12
0.80
0.31
0.29
0.67
0.69
0.93
1.11
1.20
1.48
1.26
1.43
1.29
1.39
0.59
Average
Flue 02r/
(%)
16.3
16.9
(16.9)
17.4
17.9
16.3
14.2
15.4
16.1
(12.6)
12.7
11.0
12.0
12.1
13.8
14.5
Average
Flue Temp5'
(°C)
241.7
219.3
204.4
208.1
247.6
153.3
232.4
206.2
154.8
246.3
274.2
305.4
274.1
249.8
180.4
285.9
OJ
-Ca
(Continued)
-------
Table 3-10 (Continued)
a/ Sampling Code—Refers to home and sampling rotation. For example, V05-4 is the fourth scheduled sampling rotation
in Vermont home 5. Sampling rotations 1 through 3 were conducted in approximately January, February, and March in
the 1985-86 heating season, while rotations 4 through 7 were conducted in approximately December, January, February,
and March in the 1986-87 heating season. Missing rotations indicate that data were unavailable, unusable, or
unacceptable (see Appendix B for Quality Assurance issues).
b/ Stove Code—Stoves were donated to the study by participating manufacturers. Study sponsors agreed to mask the
identity of the stoves in exchange for the generosity of the donors. Commercial use of study results by stove
manufacturers requires prior approval by the project sponsors.
c/ Sampling Period—The period during which the AWES system was collecting sample. For virtually all samples, this
was from 0000 hours Sunday through 2355 hours Saturday. All wood weights, heating degree data, etc. correspond with
this period.
d/ HDD--Heating degree days during the sampling period. Weather data was recorded in Vermont at Waterbury, and in
New York at Glens Falls by the Northeast Regional Climate Center.
e/ Catalyst Operation (%)--Defined as the percent of time the catalyst was in operation (>260°C) while the stove was
in operation (flue gas temperature >38°C). Absolute combustor temperature was used instead of catalyst AT due to
temperature measurement anomalies caused by some stove designs.
f/ Stove Operation (5)—Defined as the percent of time during the sampling period that the flue gas temperature was
>38°C. Flue gas temperature was measured approximately 45 cm above the flue collar of the stove or add-on device.
9/ Heating System Use (%)—Defined as the percent of time during the sampling period that an alternate heat source
was used in the room with the stove while the stove was operational (flue gas temperature >38°C). A thermal sensor
with a trip point of 35°C was placed in the vent of forced-air central heaters and base-board electric heaters.
n/ Efficiency (%)--Overall thermal efficiency of the stove, calculated using a modified version developed by the
Condar Company. However, flue gas temperatures were measured lower in the flue system, resulting in higher flue
temperatures and lower efficiencies. These values should be considered consistently low and for general comparison
use only. High particulate values on some samples required extrapolation.
i/ Fuel Moisture (% DB)--Fuel moisture on a dry basis (DB), as measured by a resistance pin meter. Measurements were
made at the beginning and end of each sampling period in fuel stacked near the stove for immediate use. Fuel
moistures above about 30% have a higher degree of uncertainty, due to limitations of measurement technology.
-------
Table 3-10 (Continued)
J' -Average Fuel Load (kg dry)--The average amount of fuel, normalized to 0% moisture, placed in the stove each time
the stove was fueled. Weights and fueling events were recorded automatically when the homeowner used the scale and
keypad provided.
k' Loading Frequency (#/hr)--The average number of stove fuelings per hour of stove operation (flue gas temperature
>38°C).
m' Particulate Emissions—Measured by AWES sampler. The AWES sampler, in comparison with EPA Method 5H, showed
comparable accuracy, especially in the particulate emission ranges measured in field testing. See Appendix for
detaiIs.
n' g/hr--Particulate emission rate in grams material per hour of stove operation (flue gas temperature >38°C).
°' g/kg--Particulate emission rate in grams material per kilogram fuel (normalized to 0% moisture) burned.
P' g/10^J--Particulate emission rate in grams material per million joule. Stove efficiency calculated using modified
Condar method (see note h). Heat content of fuel based on reference values for individual wood species burned.
Q' g/m3--Concentration of particulate material collected by sampler. Normalized by periods of stove operation (flue
gas temperature >38°C). Sampling was conducted at the collar of stoves or exit of add-on devices.
r/ Average Flue 02 (%)--Average concentration of oxygen in flue gas, measured by an electrochemical cell in the AWES
sampler, during periods of stove operation (flue gas temperature >38°C).
s' Average Flue Temperature (°C)--Average flue gas temperature 30 cm above stove flue collar or add-on exit during
periods of stove operation (flue gas temperature >38°C).
^•1 Catalyst was improperly seated during sampling, allowing flue gas to pass around, as well as through, catalyst.
Data not used in averaging or summaries.
u/
Averaged results from 2 AWES units sampling simultaneously.
v/ 02 > ±2% absolute at final calibration or during no-burn period. Calculated emissions (g/hr, g/kg, g/10°J)
therefore have a higher degree of associated uncertainty. Concentration (g/nr) and burn rate (kg/hr) are unaffected.
Affected values not used in averaging or summaries.
w/ Only one homeowner was weighing wood for this sampling period. All parameters except concentration (g/m^)
affected. Affected values not used in averaging or summaries.
-------
Table 3-10 (Continued)
x/ Combustor was replaced with "Long Life" catalyst for second heating season.
y' Zoned electric baseboard heat used in this home. Heating system use (%) reflects use of the heater in the room
with the stove. Undocumented use of uninstrumented baseboard heaters outside the stove room may have occurred.
z' Catalyst temperatures and field observations indicate that the combustor was less active/not active during the
second heating season.
aa' Homeowner replaced combustor between first and second year.
Low average 0_2 (8.8%) results in low calculated emission rates. Other samples from this home had average 02 of
about 16% at similar burn rates. Particulate concentration is not remarkably low. Data not used in averaging or
summaries.
cc/ Catalyst found to be damaged after this sampling period; center of combustor had dropped out. Combustor was
replaced. Note lower emission rates before catalyst failure and after replacement. Data not used in averaging or
summaries.
Solar alternate heat source used in this home. Had difficulty setting trip point of thermal sensor; recorded
alternate heat use percentages are probably lower than actual heat use percentages.
ee/ Catalyst thermocouple failed; catalyst operation (%) not calculated.
Failed combustor (substrate deterioration) discovered after this sampling period. Data not used in averaging or
summaries.
99/ Different add-on device was used for rotations V10-5 and V10-6 than for V10-2.
Add-on by-pass lever was being left in partial by-pass mode during operation period for V10-5. Homeowner was
instructed to fully close by-pass lever during V10-6 sampling period.
ii/ Different add-on device used for N04-5 than for N04-1 and N04-6.
Different add-on device used for N14-4 and N14-5 than for N14-2.
Homeowner installed flue damper. Used periodically to "hold coals" at end of burn and not typically used during
burning periods.
-------
period dates, heating degree-days (Fahrenheit basis), catalyst operation (%), stove
operation (%), alternate heating system use (%), and overall woodstove efficiency
(%). Data presented in Table 3-10B include fuel moisture (% dry basis), average
fuel load (dry kg), fuel loading frequency (#/hr), and burn rate (kg/hr). Data
presented in Table 3-10C include burn rate (kg/hr), particulate emission.rates
(g/hr, g/kg, g/106 joule, and g/m3), average flue oxygen (%), and average flue gas
temperature (°C).
The data presented in Tables 3-10A, 3-10B, and 3-10C include only results with a
high degree of confidence. "Atypical" results are shown (in parentheses, with
explanations), but are not included in data summaries or figures. The data from
these tables form the basis for the majority of the analyses undertaken in this
report. Figures 3-6A through 3-6D show the gram-per-hour emission rates measured
in the Group I and Group III homes during individual sampling periods. Figure 3-7A
through 3-7D show the burn rates (kg/hr) measured in the Group I and Group III
homes during individual sampling periods.
Tables 3-11A and 3-1IB summarize several data columns from Tables 3-10A, 3-10B, and
3-10C by stove code. Each stove code subsection contains data for homes which used
that particular stove. Table 3-11A contains data on catalyst operation time (%)
where applicable, average fuel load (dry kg), and fuel loading frequency (#/hr).
Table 3-1IB contains data on particulate emissions (g/hr and g/kg) and burn rate
(kg/hr). Overall means, standard deviations, ranges of values, and sample
populations are presented for the parameters in the tables.
Figure 3-8 is a bar graph showing the mean particulate emissions (g/hr) by
individual stove model for all stoves evaluated in the study. Figures 3-9 and 3-10
graph the overall mean particulate emission rates (g/hr for Figure 3-9, g/kg for
Figure 3-10) by stove technology type.
Although the stoves in this study are compared by technology group, it should be
remembered that these units were provided to the study and do not necessarily
represent the typical performance of any stove technology.
Catalyst Operational Time
Catalyst operational time was examined to evaluate the frequency of catalytic
activity in catalyst-equipped stoves, retrofits, and add-ons. Defined as the
percentage of time the catalyst was operational (in-catalyst temperature greater
than 380°C [500°F]) while the stove was operational (flue gas temperature greater
3-68
-------
GO
I
cn
50 -
fS -
to -
PflRTICULflTE
EMISSIONS H
(6/Hft)
30 -
25 -
20 -
15 -
10 -
5 -
Figure 3-6A
Particulate Emissions (g/hr):Individual Sampling Periods
Catalytic Stoves
N01
(fl)
N10
(fl)
U05
(B)
Ull
(B)
N03
(B)
N18
(B)
U07
(C)
U16
(C)
N03
(C)
1985-1986 HERTIHG SEftSDN
1985-13S7 HEfiTING SERSDN
-------
Figure 3-6A (Continued)
Particulate Emissions (g/hr):Individual Sampling Periods
Catalytic Stoves
50 H
to H
PflRTICULfiTE
EMISSIONS
(5/HR)
30 H
20 H
10 H
5 H
U08
(D)
U13
(D)
1985-198S HEfiTIMG SEflSDM
1385-1987 HEflTIHG SEflSDH
H02
CD)
Nil
(D)
U31
(P)
U32
(P)
H32
(P)
H33
(P)
-------
CO
50 -
tS -
to -
PflRTICULflTE
EMISSIONS H
(Q/HR)
30 -
25 -
20 -
15 -
10 -
5 -
Figure 3-6B
Particulate Emissions (g/hr):Individual Sampling Periods
Add-On/Retrofits
U03
-------
Figure 3-6C
Participate Emissions (g/hr):Individual Sampling Periods
Low-Emission Stoves
t5 -
to -
PflRTICULflTE
EMISSIONS
(3/HR)
30 -
CO
1
20 -
10 -
5 -
U18
CK)
N07
(K)
not
(L)
HIS
(L)
U12
(M)
Ult
CM)
U3t
Cfl)
U03
(H)
U35
(H)
N16
(H)
1385-1986 HEflTIHG SEflSDH
1986-1387 HEfiTING SEflSDH
-------
50
f 5
PflRTICULfiTE
EMISSIONS
30
25
20
15
10
S -
Figure 3-6 D
Particulate Emissions (g/hr):Individual Sampling Periods
Conventional Stoves
U09
(D)
1985-1986 HEflTING SEfiSQN
1986-1987 HERTING SEflSQN
Ulf
(D)
N08
(D)
Nit
(D)
N16
(D)
-------
2.5-
6 URN
RftTE
(KG/HR)
i.5 -
1.0 -
0,5 -
Figure 3-7A
Burn Rate (kg/hr):Individual Sampling Periods
Catalytic Stoves
A
HOI
(fl)
H10
(fl)
U05
(B)
Ull
(B)
H09
(B)
N18
(B)
U07
(C)
U16
(C)
1985-1986 HEflTING SEflSDN
1386-1987 HEflTING SEflSDN
N03
(C)
-------
Figure 3-7A (Continued)
Burn Rate (kg/hr):Individual Sampling Periods
Catalytic Stoves
3.0 -
2.5 -
BURN
RftTE
(KG/HR)
1.5 -
1.0 -
0.5 -
U08
(D)
U13
CD)
1985-1986 HEflTING SERSQN
1986-1987 HEflTING SEflSQN
N02
(D)
Nil
(D)
U31
(P)
U32
(P)
N32
(P)
N33
(P)
-------
Figure 3-7B
Burn Rate (kg/hr):Individual Sampling Periods
Add-On/Retrofits
3.0 -
2.5 -
BURN
RATE -
(KG/HR)
l.S -
i.O -
0.5 -
UOi
(E)
U03
(F)
U12
CF)
U02
CG)
N04
(G)
U10
(H)
N06
( I)
Hit-
( I)
U10
(J)
HOt
(J)
N12
(J)
1985-1986 HEflTIHG SEflSDH
138G-1987 HEflTIHG SEflSDN
-------
Figure 3-7C
Burn Rate (kg/hr):Individual Sampling Periods
Low-Emission Stoves
3.0 -
2.5 -
BURN
RflTE H
(KQ/HR)
1,5 -
i.O -
0.5 -
U18
CK)
N07
(K)
UOt
(L)
N15
(L)
U12
(H)
Ult
CH>
CM)
U03
CN)
U35
(H)
Hie
(N)
1985-1986 HERTING SEflSDN
198S-1987 HEfiTING SEftSDN
-------
Figure 3-7D
Burn Rate (kg/hr):Individual Sampling Periods
Conventional Stoves
3.0-
2.5-
BURN
RftTE H
(KG/HR)
1.5 -
1.0 -
0.5 -
U06
(D)
U09
(D)
1985-1986 HEflTIHG SEflSDN
1386-1987 HEflTIHG SEflSOH
U1H
(D)
N08
(Q)
Hit
(D)
H16
(D)
-------
Table 3-11A
STOVE USE CHARACTERISTICS BY STOVE MODEL -- CATALYTIC STOVES
Stove
Code3/
A
B
C
D
P
Firebox
Volume
(liters)
87
122
69
38
40
52
87
119
N/A
ALL CATALYTIC
STOVES
Home
Code
N01
N10
ALL
V05
Vll
N09
N18
ALL
V07
V16
N03
ALL
V08
V13
N02
Nil
ALL
V31
V32
N32
N33
ALL
N/A
Catalyst Operation'3/
(percent)
Mean
26.0
86.3
58.9
88.0
54.1
77.2
87.7
76.7
71.9
63.5
47.5
63.6
59.4
45.5
66.3
46.7
54.9
87.8
40.9
64.1
58.8
59.4
62.1
,<*/
17.0
3.9
32.3
0.2
14.1
8.7
2.6
15.7
9.5
12.2
7.7
13.8
9.1
8.1
10-. 7
13.4
13.2
0
7.1
12.8
0
18.1
20.3
Range
10.8-57.0
80.3-91,1
10.8-91.1
87.8-88.1
42.9-74.5
67.3-88.2
85.0-90.9
42.9-90.9
55.7-85.4
43.2-80.2
41.4-58.4
41.4-85.4
40.9-70.9
31.8-57.4
53.9-83.1
31.7-65.0
31.7-83.1
33.8-48.0
51.3-76.8
33.8-87.8
10.8-91.1
Nf/
5
6
11
2
3
4
4
13
6
5
3
14
7
6
5
4
22
1
2
2
1
6
66
Average Fuel Load0/
(kilograms dry)
Mean
4.5
7.2
6.0
5.2
12.7
10.6
7.5
9.3
9.1
6.4
5.5
7.2
5.0
4.0
5.4
2.7
4.4
2.9
5.3
7.0
6.9
5.9
6.4
,e/
0.2
2.1
2.0
0.8
2.7
1.1
1.1
3.0
0.7
0.6
0.2
1.6
0.7
0.4
0.4
0.2
l.l_l
0
1.2
0.5
0.1
1.6
2.6
Range
4.3- 4.9
4.7-11.1
4.3-11.1
4.4- 6.0
9.0-15.1
8.8-11.6
5.9- 9.1
4.4-15.1
8.4-10.2
5.5- 7.5
5.3- 5.7
5.3-10.2
3.8- 5.7
3.2- 4.4
4.8- 5.8
2.4- 2.9
2.4- 5.8
4.1- 6.4
6.5- 7.4
6.8- 6.9
2.9- 7.4
2.4-15.1
Nf/
5
6
11
2
3
4
4
13
5
5
3
13
7
6
5
4
22
1
2
2
2
7
66
Loading Frequency^/
(number per hour)
Mean
0.15
0.21
0.18
0.17
0.09
0.12
0.19
0.14
0.16
0.22
0.18
0.19
0.20
0.27
0.19
0.33
0.24
0.35
0.23
0.17
0.30
0.25
0.20
,e/
0.01
0.05
0.05
0.02
0.03
0.01
0.01
0.04
0.04
0.06
0.01
0.05
0.02
0.03
0.02
0.07
0.06
0
0.05
0.01
0.03
0.07
0.07
Range
0.13-0.17
0.14-0.28
0.13-0.28
0.15-0.19
0.07-0.13
0.11-0.14
0.17-0.20
0.07-0.20
0.11-0.22
0.17-0.35
0.17-0.19
0.11-0.35
0.16-0.21
0.22-0.30
0.15-0.22
0.22-0.40
0.15-0.40
0.18-0.27
0.16-0.17
0.27-0.33
0.16-0.35
0.07-0.40
Nf/
5
6
11
2
3
4
4
13
5
6
3
14
7
6
5
4
22
1
2
2
2
7
66
(Continued,
00
I
-------
Table 3-11A (Continued)
STOVE USE CHARACTERISTICS BY STOVE MODEL -- ADD-ON/RETROFIT DEVICES^/
Stove
Code3/
E(R)
F(R)
G(A)
H(A)
KA)
J(A)
Firebox
Volume
(liters)
62
74
74
N/A
77
84
N/A
81
102
N/A
78
119
N/A
81
84
111
119
N/A
ALL (R)
ALL (A)
ALL ADD-ON/
RETROFITS
Home
Code
V01
(ALL)
V03
V12
ALL
V02
N04
ALL
V10
V15
ALL
N06
N14
ALL
V10
N04
N12
N14
ALL
N/A
Catalyst Operationb/
(percent)
Mean
69.3
21.6
9.8
17.7
43.5
27.9
35.7
19.0
22.4
20.7
58.7
53.7
57.5
36.9
57.8
53.6
68.2
50.7
43.5
44.5
48.4
,e/
4.4
3.9
0
6.4
6.0
18.5
15.8
0
0
1.7
5.7
0
5.4
19.3
0
0
0
17.3
26.4
18.1
22.3
Range
63.9-74.7
17.7-25.5
9.8-25.5
37.5-49.5
9.4-46.4
9.4-49.5
19.0-22.4
53.2-66.6
53.2-66.6
17.6-56.2
17.6-68.2
9.8-74.7
9.4-68.2
9.4-74.7
Nf/
3
2
1
3
2
2
4
1
1
2
3
1
4
2
1
1
1
5
6
15
21
Average Fuel Loadc/
(kilograms dry)
Mean
6.1
6.5
4.2
5.1
11.3
8.1
9.7
4.0
0
4.0
8.2
6.9
7.9
3.9
7.9
6.9
6.0
5.7
5.5
7.3
6.6
,*l
0.5
1.2
0.5
1.4
0.8
0.2
1.7
0
0
0
0.8
0
0.9
0.1
0
0
0
1.6
1.2
2.3
2.1
Range
5.3- 6.4
5.3- 7.6
3.9- 4.9
3.9- 7.6
10.5-12.0
7.9- 8.2
7.9-12.0
7.3- 9.2
6.9- 9.2
3.8- 4.0
3.8- 7.9
3.9- 7.6
3.8-12.0
3.8-12.0
Nf/
4
2
3
5
2
2
4
1
0
1
3
1
4
2
1
1
1
5
9
14
23
Loading Frequency"'
(number per hour)
Mean
0.22
0.19
0.22
0.20
0.14
0.22
0.18
0.23
0
0.23
0.28
0.34
0.29
0.33
0.21
0.19
0.30
0.27
0.21
0.25
0.23
,*l
0.02
0.03
0.07
0.06
0.01
0.01
0.04
0
0
0
0.03
0
0.04
0.05
0
0
0
0.06
0.05
0.07
0.06
Range
0.21-0.26
0.16-0.21
0.12-0.28
0.12-0.28
0.13-0.15
0.21-0.23
0.13-0.23
0.25-0.32
0.25-0.34
0.28-0.37
0.19-0.37
0.12-0.28
0.13-0.37
0.12-0.37
Nf/
4
2
3
5
2
2
4
1
0
1
3
1
4
2
1
1
1
5
9
14
23
(Continued
CO
o
-------
Table 3-11A (Continued)
STOVE USE CHARACTERISTICS BY STOVE MODEL — LOW-EMISSION STOVES
co
i
00
Stove
Code3/
K
L
M
N
Firebox
Volume
(liters)
37
37
41
49
ALL LOW-
EMISSION STOVES
Home
Code
V18
N07
ALL
V04
N15
ALL
V12
V14
V34
N13
ALL
V03
V35
N16
ALL
N/A
Average Fuel Load0'
(kilograms dry)
Mean
3.6
5.0
4.2
2.7
2.6
2.6
2.2
3.6
3.3
3.9
3.3
4.9
3.1
3.1
3.7
3.5
.*/
0.3
0.5
0.8
0.2
0.3
0.2
0
0.1
0.5
0
0.6
0.3
0
0.2
0.9
0.9
Range
3.1- 4.0
4.6- 5.7
3.1- 5.7
2.5- 2.8
2.2- 2.9
^ 2.2- 2.9
3.5- 3.7
2.8- 3.8
2.2- 3.9
4.6- 5.2
2.9- 3.3
2.9- 5.2
2.2- 5.7
Nf/
4
3
7
4
3
7
1
2
2
1
6
2
1
3
6
26
Loading Frequency0*/
(number per hour)
Mean
0.32
0.18
0.26
0.34
0.45
0.38
0.30
0.27
0.26
0.31
0.28
0.27
0.29
0.31
0.30
0.30
,*/
0.02
0.02
0.07
0.03
0.09
0.09
0
0.03
0.02
0
0.03
0.03
0
0.03
0.03
0.08
Range
0.29-0.35
0.15-0.20
0.15-0.35
0.31-0.38
0.32-0.53
0.31-0.53
0.24-0.29
0.24-0.27
0.24-0.31
0.24-0.30
0.27-0.34
0.24-0.34
0.15-0.53
Nf/
4
3
7
4
3
7
1
2
-.2
1
6
2
1
3
6
26
(Continued,
-------
Table 3-11A (Continued)
STOVE USE CHARACTERISTICS BY STOVE MODEL -- CONVENTIONAL TECHNOLOGY STOVES
Stove
Code3/
0
Firebox
Volume
(liters)
84
77
65
84
119
33
ALL CONVEN-
TIONAL STOVES
Home
Code
V06
V09
V14
N08
N14
N16
N/A
Average Fuel Load0'
(kilograms dry)
Mean
7.2
3.8
4.3
8.3
5.5
3.5
6.2
,e/
0.5
0
0.6
1.0
0.1
0
1.8
Range
6.7- 8.0
3.5- 4.7
7.2- 9.8
5.4- 5.6
3.5- 9.8
Nf/
5
1
3
4
3
1
17
Loading Frequency0'/
(number per hour)
Mean
0.25
0.30
0.31
0.25
0.37
0.45
0.30
„*/
0.03
0
0.04
0.03
0.06
0
0.07
Range
0.22-0.31
0.26-0.35
0.20-0.28
0.29-0.43
0.20-0.45
Nf/
5
1
3
4
3
1
17
Co
ro
-------
Table 3-11B
EMISSION AND BURN RATE CHARACTERISTICS BY STOVE MODEL — CATALYTIC STOVES
GO
I
00
oo
Stove
Code3/
A
B
C
D
P
Home
Code
N01
N10
ALL
V05
Vll
N09
N18
ALL
V07
V16
N03
ALL
V08
V13
N02
Nil
ALL
V31
V32
N32
N33
ALL
ALL CAT.
STOVES
Participate EmissionsQ/
(grams per hour)
Mean
18.0
22.8
20.4
20.2
6.5
20.9
30.7
20.5
9.4
16.2
17.1
14.2
14.4
13.7
8.4
10.2
12.2
17.7
12.9
19.6
28.5
20.0
16.4
Si
3.7
10.9
8.5
11.2
0.4
5.4
7.4
11.1
4.7
4.6
6.7
6.2
4.1
3.6
1.3
4.7
4.3
0
1.1
0
6.2
7.4
8.4
Range
13.0-21.9
9.7-39.7
13.0-39.7
9.0-31.4
6.1- 7.0
15.7-29.6
20.6-41.3
6.1-41.3
1.7-14.3
8.2-21.8
8.1-24.3
1.7-24.3
7.6-20.4
9.7-19.4
6.9-10.0
5.5-14.9
5.5-20.4
11.8-13.9
22.3-34.6
11.8-34.6
1.7-41.3
Nf/
4
4
8
2
3
4
4
13
4
5
3
12
6
6
5
2
19
1
2
1
2
6
58
Particulate Emissions'1/
(grams per kilogram)
Mean
25.5
14.5
20.0
22.5
5.8
16.1
22.2
16.6
6.2
13.4
17.1
12.0
14.2
13.4
8.2
11.4
12.1
17.5
11.4
16.6
13.8
14.1
14.4
Si
2.9
6.9
7.6
11.7
0.4
3.4
4.7
8.5
3.4
4.0
6.6
6.3
2.2
6.5
0.7
5.4
4.9
0
1.1
0
1.6
2.6
7.0
Range
22.2-28.6
6.7-25.1
6.7-28.6
10.8-34.1
5.3- 6.1
12.8-21.5
15.1-26.5
5.3-34.1
1.2-10.5
6.3-17.4
8.1-23.6
1.2-23.6
11.7-17.8
8.1-26.7
6.7- 8.7
6.0-16.7
6.0-26.7
10.3-12.4
12.2-15.3
10.3-17.5
1.2-34.1
Nf/
4
4
8
2
3
4
4
13
4
b
3
12
6
6
5
2
19
1
2
1
2
6
58
Burn Rate1/
(kilograms per hour)
Mean
0.70
1.42
1.10
0.88
1.12
1.29
1.38
1.21
1.60
1.21
1.00
1.31
1.00
1.11
1.04
0.89
1.02
1.01
1.13
1.14
2.05
1.38
1.17
Si
0.07
0.20
0.40
0.04
0.07
0.06
0.13
0.19
0.22
0.08
0.02
0.28
0.22
0.17
0.13
0.21
0.20
0
0.01
0.05
0.22
0.44
0.32
Range
0.57-0.78
1.16-1.69
0.57-1.69
0.84-0.92
1.02-1.19
1.23-1.37
1.19-1.57
0.84-1.57
1.35-1.89
1.11-1.29
0.97-1.03
0.97-1.89
0.61-1.27
0.73-1.24
0.82-1.17
0.58-1.16
0.58-1.27
1.12-1.14
1.09-1.18
1.83-2.26
1.01-2.26
0.57-2.26
Nf/
5
6
11
2
3
4
4
13
5
5
3
13
7
6
5
4
22
1
2
2
2
7
66
(Continued,
-------
Table 3-11B (Continued)
EMISSION AND BURN RATE CHARACTERISTICS BY STOVE MODEL -- ADD-ON/RETROFIT DEVICES^/
OJ
I
CO
Stove
Code9/
E(R)
F(R)
G(A)
H(A)
KA)
J(A)
Home
Code
V01
(ALL)
V03
V12
ALL
V02
N04
ALL
V10
(ALL)
N06
N14
ALL
V10
N04
N12
N14
ALL
ALL (R)
ALL (A)
ALL ADD-ON/
RETROFITS
Particulate Emissions9/
(grams per hour)
Mean
7.8
25.2
26.6
25.9
16.3
18.7
17.1
16.2
22.6
25.7
23.4
14.9
14.2
7.3
k
12.8
18.2
17.7
17.9
,e/
1.6
6.6
10.1
8.6
0.8
0
1.3
0
10.5
0
9.2
6.5
0
0
k
5.6
11.1
7.6
9.3
Range
6.3-10.1
18.6-31.8
16.5-36.7
16.5-36.7
15.5-17.1
15.5-18.7
13.6-37.3
13.6-37.3
8.4-21.3
k
7.3-21.3
6.3-36.7
7.3-37.3
6.3-37.3
Nf/
3
2
2
4
2
1
3
1
3
1
4
2
1
1
k
4
7
12
19
Particulate Emissions"/
(grams per kilogram)
Mean
6.0
24.1
27.1
25.8
10.1
11.0
10.4
16.1
10.4
10.9
10.5
12.8
8.6
5.5
k
4
17.3
10.7
13.2
.*!
1.0
12.4
10.3
11.6
0.5
0
0.6
0
5.4
0
6.0
7.1
0
0
k
5.9
13.1
4.6
9.6
Range
5.3- 7.5
11.7-36.5
17.1-37.9
11.7-37.9
9.6-10.5
9.6-11.0
5.9-17.9
5.9-17.9
5.7-19.9
k
5.5-19.9
5.3-37.9
5.5-19.9
5.3-37.9
Nf/
3
2
2
4
2
1
3
1
3
1
4
2
1
1
k
4
7
12
19
Burn Rate1/
(kilograms per hour)
Mean
1.37
1.23
1.10
1.15
1.62
1.78
1.70
1.01
2.23
2.35
2.26
1.27
1.66
1.31
1.78
1.46
1.25
1.75
1.56
,e/
0.15
0.36
0.19
0.28
0.01
0.08
0.10
0
0.11
0
0.11
0.20
0
0
0
0.25
0.25
0.42
0.44
Range
1.17-1.58
0.87-1.59
0.97-1.37
0.87-1.59
1.61-1.62
1.70-1.86
1.61-1.86
2.08-2.32
2.08-2.35
1.07-1.47
1.07-1.78
0.87-1.59
0.17-2.35
0.87-2.35
Nf/
4
2
3
5
2
2
4
1
3
1
4
2
1
1
1
5
9
15
24
(Continued,
-------
Table 3-11B (Continued)
EMISSION AND BURN RATE CHARACTERISTICS BY STOVE MODEL — LOW-EMISSION STOVES
Stove
Code3/
K
L
M
N
Home
Code
V18
N07
ALL
V04
N15
ALL
V12
V14
V34
N13
ALL
V03
V35
N16
ALL
ALL L.E.
STOVES
Particulate Emissions9/
(grams per hour)
Mean
29.5
11.2
23.4
9.2
9.6
9.4
5.2
21.8
6.9
k
12.5
10.2
3.6
8,2
8.1
13.4
„*/
11.2
1.8
12.6
3.5
1.4
2.6
0
4.6
1.0
k
8.1
8.2
0
2.8
5.5
10.2
Range
17.3-47.6
9.4-12.9
9.4-47.6
6.5-14.1
7.9-11.4
6.5-14.1
17.2-26.3
5.9- 7.9
k
5.2-26.3
2.0-18.3
4.3-10.3
2.0-18.3
2.0-47.6
Nf/
4
2
6
3
3
6
1
2
2
k
5
2
1
3
6
23
Particulate Emissions'1/
(grams per kilogram)
Mean
25.6
12.9
21.4
11.4
8.7
10.0
7.7
22.5
8.4
k
13.9
7.9
4.0
8.7
7.6
13.2
**l
8.0
2.5
9.0
5.0
2.6
4.2
0
2.1
2.0
k
7.3
6.5
0
3.5
4.8
8.6
Range
16.0-37.9
10.4-15.3
10.4-37.9
7.7-18.4
5.9-12.2
5.9-18.4
20.4-24.6
6.4-10.4
k
6.4-24.6
1.4-14.3
3.9-11.9
1.4-14.3
1.4-37.9
Nf/
4
2
6
3
3
6
1
2
2
k
5
2
1
3
6
23
Burn Rate1/
(kilograms per hour)
Mean
1.13
0.86
1.02
0.90
1.15
1.01
0.67
0.96
0.84
1.18
0.91
1.33
0.90
0.98
1.08
1.00
,e/
0.07
0.03
0.15
0.11
0.17
0.15
0
0.11
0.08
0
0.17
0.05
0
0.09
0.19
0.19
Range
1.08-1.26
0.84-0.90
0.84-1.26
0.76-1.07
0.93-1.34
0.76-1.34
0.85-1.07
0.76-0.92
0.67-1.18
1.28-1.38
0.87-1.10
0.87-1.38
0.67-1.38
Nf/
4
3
/
4
3
7
1
2
2
1
6
2
1
3
6
26
(Continued
CO
en
-------
Table 3-11B (Continued)
EMISSION AND BURN RATE CHARACTERISTICS BY STOVE MODEL -- CONVENTIONAL TECHNOLOGY STOVES
Stove
Code3/
0
Home
Code
V06
V09
V14
N08
N14
N16
ALL CONV.
STOVES
Particulate EmissionsS/
(grams per hour)
Mean
9.4
15.4
20.2
30.0
31.5
13.9
20.1
,e/
5.9
0
2.7
2.5
2.5
0
9.5
Range
2.9-17.3
16.9-23.5
26.6-32.6
29.0-34.0
2.9-34.0
Nf/
4
1
3
3
2
1
14
Particulate Emissions"/
(grams per kilogram)
Mean
5.5
13.7
16.2
14.9
16.2
9.0
12.1
c*l
3.5
0
4.8
2.0
2.3
0
5.6
Range
1.2- 9.3
10.2-22.0
12.2-17.1
13.9-18.4
1.2-22.0
Nf/
4
1
3
3
2
1
14
Burn Rate"'/
(kilograms per hour)
Mean
1.80
1.12
1.35
2.01
2.06
1.55
1.76
.*!
0.34
0
0.31
0.11
0.37
0
0.41
Range
1.52-2.45
1.45-1.92
1.91-2.19
1.57-2.45
1.45-2.45
Nf/
5
1
3
4
3
1
17
en
CTl
-------
Table 3-11
STOVE USE AND EMISSION AND BURN RATE CHARACTERISTICS
a' Stove Code--Stoves were donated to the study by participating manufacturers. Study sponsors agreed to mask the
identity of the stoves in exchange for the generosity of the donors. Commercial use of study results by stove
manufacturers requires prior approval by the project sponsors.
b/ Catalyst Operation (%)--Defined as the percent of time the catalyst was in operation (>260°C) while the stove was
in operation (flue gas temperature >38°C). Absolute combustor temperature was used instead of catalyst AT due to
temperature measurement anomalies caused by some stove designs.
c' Average Fuel Load (kg dry)--The average amount of fuel, normalized to 0% moisture, placed in the stove each time
the stove was fueled. Weights and fueling events were recorded automatically when the homeowner used the scale and
keypad provided.
d/ Loading Frequency (#/hr)--The average number of stove fuel ings per hour of stove operation (flue gas temperature
>38°C).
.
e' Standard deviation.
f' Number of values.
9/ Particulate Emissions (g/hr)--Particulate emission rate in grams material per hour of stove operation (flue gas
temperature >38°C).
n/ Particulate Emissions (g/kg)--Particulate emission rate in grams material per kilogram fuel (normalized to 0%
moisture) burned.
i/ Burn Rate (kg/hr)--Average fuel consumption (normalized to 0% moisture) per hour of stove operation (flue gas
temperature >38°C).
J/ Add-on/Retrofits--(R) refers to internal catalytic retrofit, (A) refers to catalytic stack add-on.
k/ No data aviTable due to sampling equipment malfunction.
-------
HEflN
PflRTICULflTE
EMISSIONS
(G/HR)
i
oo
10 -
STDUE CODE:
tt OF SftMPLES:
if DF HDhES;
Figure 3-8
Particulate Emissions (g/hr) by Stove Model
E
13
C
12
D P
19 6
t H
E F G
3 t 3
i 2 2
K L
6 6
2 2
D
If
5
CflTflLVTIC STDUES
Y/////X fiDO-ON/RETRQFnS
LDH-EfllSSION STDUES
CDHUENTIDHflL STDUES
-------
Co
10
30.0 -
25.0 -
HEAN
PftRTICULATE
EMISSIONS
(G/HR)
15.0 -
10 .0 -
5.0
Figure 3-9
Particulate Emissions (g/hr) by Stove Technology
CATftLYTIC
STDUES
ADD-DH/
REIRDFITS
v * * * v v
W.V.+.V.+
*v»»v*«»
LDH-
EHISSIDN
STDUES
CDNUENTIDNfiL
STDUES
-------
2E.O H
HEfiH
F'MftTICULftTE
EMISSIONS
(G/KG)
1E.O H
1C.0 -\
5.0 H
Figure 3-18
Particulate Emissions (g/kg) by Stove Technology
CATfiLVTIC
STDUES
ADD-DM/
RETROFITS
****«
tVAV+VA
LDH-
EHISSIDH
STDUES
COHUEHTIDHflL
STDUES
-------
than 38°C [100°F]). (Using temperature increase [AT] across the combustor as an
indicator of catalyst activity was evaluated, but not used due to interferences
caused by some stove designs and by some thermocouple installations. For example,
in some cases the firebox (before catalyst) thermocouple was located downstream of
the point of secondary air entry causing false AT values under some burning
conditions. A number of AT values were negative. Table 3-12 presents AT data.)
Catalyst temperatures, percent operation time, AT values, and bypass usage figures
did not provide an explanation of high or low emission rate trends. Results were
scattered and inconsistent. Conditions which would be expected to result in high
emissions would have low emissions and vice-versa.
The overall mean catalyst operational time presented in Table 3-11A for the
catalytic stoves (62,1%) was approximately 28% higher than the overall mean
catalyst operational percentage for the add-on/retrofits (48.4%). This apparent
difference could be either an actual difference due to higher temperatures
generated by the catalytic stoves or an artifact of the catalyst thermocouple probe
placement on the add-on/retrofit devices. For the catalytic stoves, thermocouples
were generally placed in the catalyst substrate, so an actual in-catalyst
temperature was measured. Due to movable catalysts in some add-on devices,
thermocouple probes were placed as closely as possible to the catalyst, but not in
the substrate. Consequently, the catalyst operational time for the add-on/
retrofits may be conservative (low) if the overall mean is actually affected by the
catalyst probe placement.
Fuel Load Data
Average fuel load data provide information on the mass of wood placed in the stove
each time the unit was fueled. Fuel load data are normalized to zero percent
moisture. The overall mean fuel loads for the catalytic stoves (6.4 kg), add-on/
retrofits (6.6 kg), and conventional stoves (6.2 kg) were all within a narrow
range. The low-emission stoves (which generally have small firebox volumes
relative to catalytic stoves and conventional stoves) had an overall mean fuel load
(3.5 kg) that was approximately 55% of the overall mean fuel loads for the
catalytic stoves, add-on/retrofits, and conventional stoves.
The loading frequency (#/hr) data is an indication of the average number of times
per hour that homeowners fueled their stoves. The inverse of this parameter is the
average time between refueling. As would be expected based on the overall mean
fuel load, the low-emission stoves had one of the highest overall mean loading
3-91
-------
Table 3-12
CATALYST OPERATIONAL CHARACTERISTICS
I
U3
fxj
Sample
Code
V01-49/
V01-59/
V01-69/
V01-79/
V05-4
V05-5
V05-6
V07-5
V07-6
V07-7
V08-4
V08-5
V08-6
V08-7
V10-5h/
V10-6
V10-7
Vll-6
Vll-7
Stove
Code
E
B
C
D
J
B
% Time
Stove
Opera-
tional3/
99.9
100.0
100.0
95.4
100.0
85.7
62.5
99.3
97.0
97.2
85.7
100.0
95.5
93.7
74.6
56.2
12.3
98.8
54.7
% Time
Catalyst
Opera-
tional13/
69.2
74.7
75.3
63.9
88.1
87.8
87.9
65.5
74.9
70.9
58.4
68.7
70.9
62.0
17.6
43.9
41.0
45.0
42.9
Average
Catalyst
Temperature
°C (°F)C/
393 (739)
442 (827)
413 (776)
333 (631)
407 (765)
401 (753)
388 (730)
327 (620)
381 (717)
368 (694)
306 (583)
336 (636)
382 (719)
343 (649)
188 (370)
261 (501)
219 (427)
298 (568)
262 (503)
Average
Catalyst
°C (°F)d/
2 (4)
16 (28)
-20 (-36)
-4 (-8)
207 (372)
197 (354)
173 (312)
-32 (-58)
-47 (-85)
-4 (-8)
4 (7)
-9 (-17)
-11 (-19)
-8 (-15)
-22 (-39)
8 (15)
23 (42)
97 (175)
82 (147)
% Time
Bypass
Open6/
1.1
1.3
1.7
15.8
0.0
0.0
0.0
3.1
2.9
5.3
0.2
0.0
0.7
3.1
0.2
0.0
0.0
1.4
1.2
Average
Flue
Temperature
°C (°F)f/
202 (395)
221 (429)
230 (446)
220 (428)
72 (161)
58 (137)
63 (145)
173 (343)
205 (401)
211 (412)
168 (334)
169 (336)
253 (487)
229 (445)
134 (273)
66 (151)
150 (303)
148 (299)
131 (267)
Particulate
Emission
Rate
(9/hr)
6.3
10.1
16.7
7.1
9.0
31.4
4.8
14.3
1.7
7.6
12.1
13.4
12.7
21.3
8.4
6.3
7.0
(Continued)
-------
Table 3-12 (Continued)
CATALYST OPERATIONAL CHARACTERISTICS
OJ
CO
Sample
Code
V13-4
V13-5
V13-6
V13-7
V16-4
V16-5
V16-6
V16-7
V31-41/
V-3251'/
N01-4
N01-5
N01-6
N01-7
N02-4
N02-5
N02-6
N02-7
N03-4
N03-5
N03-6
N03-7
Stove
Code
D
C
P
P
A
D
C
% Time
Stove
Opera-
tional3/
99.8
100.0
100.0
96.3
93.6
70.7
87.0
71.6
100.0
100.0
97.1
100.0
100.0
98.8
98.2
100.0
100.0
100.0
40.6
76.5
66.4
46.2
% Time
Catalyst
Opera-
tional13/
39.9
45.0
49.2
49.6
62.7
60.7
70.5
43.2
87.8
48.0
21.4
22.0
10.9
10.8
60.7
64.5
59.9
53.9
41.4
58.4
42.7
28.8
Average
Catalyst
Temperature
°C (°F)C/
247 (476)
259 (499)
283 (542)
288 (550)
313 (595)
313 (596)
339 (643)
248 (479)
463 (865)
256 (492)
206 (403)
202 (395)
181 (358)
193 (380)
331 (628)
322 (612)
303 (577)
296 (564)
256 (493)
303 (578)
251 (483)
201 (394)
Average
Catalyst
°C (°F)d/
-72(-129)
-88(-159)
-31 (-56)
-39 (-71)
19 (35)
11 (19)
-2 (-3)
-46 (-82)
--
__
-18 (-32)
-36 (-64)
-42 (-75)
-44 (-79)
-12 (-21)
-119(-215)
-155(-279)
-96(-172)
8 (14)
19 (34)
11 (20)
-18 (-33)
% Time
Bypass
Open6/
2.2
1.3
2.6
4.4
0.9
0.7
0.2
2.5
0.8
2.3
0.3
1.5
1.5
0.5
0.0
0.3
0.3
0.2
0.5
0.4
0.0
1.6
Average
Flue
Temperature
°C (°F)f/
203 (398)
198 (389)
194 (382)
199 (391)
191 (375)
182 (359)
199 (391)
191 (375)
168 (335)
140 (284)
108 (226)
125 (257)
109 (229)
121 (249)
214 (418)
262 (503)
257 (494)
248 (478)
144 (292)
143 (289)
136 (277)
126 (259)
Part icu late
Emission
Rate
(g/hr)
12.4
9.7
12.4
10.5
19.0
21.8
16.8
15.1
17.7
11.8
21.9
21.2
15.9
6.9
10.0
8.4
8.1
19.0
24.3
(Continued)
-------
Table 3-12 (Continued)
CATALYST OPERATIONAL CHARACTERISTICS
Sample
Code
N04-5
N09-4
N09-6
N09-7
N10-4
N10-5
N10-6
N10-7
Nll-4
Nll-6
Nll-7
N12-4
N12-5
N14-4
N18-4
N18-5
N18-6
N18-7
N32-5
Stove
Code
J
B
A
D
J
J
B
P
% Time
Stove
Opera-
tional3/
96.1
100.0
99.0
84.8
100.0
100.0
100.0
100.0
93.3
100.0
52.1
93.9
85.5
100.0
100.0
100.0
78.6
98.1
100.0
% Time
Catalyst
Opera-
tional5/
57.8
83.1
70.1
67.3
82.3
90.3
86.6
87.2
53.7
65.0
36.2
53.6
47.2
68.2
85.1
85.0
89.6
90.9
51.3
Average
Catalyst
Temperature
°C (°F)C/
280 (536)
348 (659)
319 (607)
295 (563)
388 (730)
411 (771)
368 (694)
349 (661)
264 (508)
307 (585)
220 (442)
280 (536)
258 (496)
284 (544)
386 (726)
392 (738)
371 (699)
394 (741)
358 (677)
Average
Catalyst
°C (°F)d/
-6 (-10)
3 (6)
3 (6)
-5 (-9)
29 (53)
25 (45)
9 (16)
19 (34)
-61(-110)
-119(-215)
-37 (-66)
17 (31)
3 (6)
38 (69)
1 (D
31 (56)
54 (98)
109 (197)
36 (64)
% Time
Bypass
Open6/
5.0
1.3
2.1
0.2
0.0
0.0
0.0
0.0
3.8
0.0
0.0
0.7
0.2
0.0
0.0
0.5
0.0
0.2
3.0
Average
Flue
Temperature
°C (°F)f/
222 (431)
263 (505)
234 (453)
111 (231)
161 (322)
184 (364)
178 (352)
164 (328)
210 (410)
92 (198)
183 (361)
205 (401)
196 (384)
229 (444)
166 (330)
170 (338)
157 (315)
206 (403)
152 (306)
Particulate
Emission
Rate
(9/hr)
14.2
21.2
17.1
29.6
13.9
23.4
18.2
39.7
5.5
6.6
4.6
7.3
15.2
20.6
41.3
31.6
29.2
19.6
(Continued)
-------
Table 3-12 (Continued)
CATALYST OPERATIONAL CHARACTERISTICS
a' Stove operational time is defined as the percentage of the sampling period in which the flue
temperature was greater than or equal to 38°C (100°F).
"' Catalyst operational time is defined as the percentage of stove operational time in which the in-
catalyst or after-catalyst temperature was greater than or equal to 260°C (500°F).
c/ Average catalyst temperature is the average of the in-catalyst or after-catalyst temperature
readings during stove operational time only.
"' Average catalyst AT is the average change in temperature of the stove exhaust gases as they pass
through the catalytic combustor. The AT is obtained by subtracting the before-catalyst thermocouple
readings from the after-catalyst or in-catalyst thermocouple readings. The average values presented
here are for only the stove operational period(s).
f e' Bypass open time is defined as the percentage of stove operational time in which the flue
cS temperature was greater than or equal to the catalyst temperature.
'' Average flue temperature is the average of flue temperature readings (measured .in the stove pipe
approximately one foot downstream of the heating appliance) during the stove operational time only.
9/ Home V01 had a damaged combustor (substrate crumbling) replaced between samples V01-6 and V01-7.
n' Homeowner of V10 was operating catalytic add-on in the partial bypass mode during sample V10-5.
Subsequent samples were collected with the add-on in full catalytic mode.
i' Before-catalyst temperature not measured in Homes V31 and V32.
-------
frequencies (0.30 #/hr). However, the conventional stoves also had an overall mean
loading frequency of 0.30 #/hr. The higher average fuel load weight, with a high
average fueling frequency for the conventional stoves, presumably indicates lower
stove efficiency. The catalytic technologies had the lowest average loading
frequencies (0.20 #/hr for catalytic stoves, 0.23 #/hr for add-on/retrofits).
Particulate Emissions
The low-emission stoves exhibited the lowest overall mean gram-per-hour emission
rate (13.4 g/hr). The catalytic technology (catalytic stoves and add-on/
retrofits) had similar mean emission rates (16.4 g/hr for catalytic stoves, 17.9
g/hr for add-on/retrofits). The conventional stoves had the highest overall mean
emission rate (20.1 g/hr).
The gram-per-kilogram emission results exhibited a different ranking by technology
classification (primarily due to variations in overall mean burn rates for the
technology classifications) than the gram-per-hour emission rates. The
conventional stoves had the lowest overall mean gram-per-kilogram emissions (12.1
g/kg). The low-emission stoves and add-on/retrofits had the same overall mean
emissions (13.2 g/kg). The catalytic stoves had the highest overall mean emissions
(14.4 g/kg). However, given the narrow range of values, these means are
statistically similar.
The low-emission stoves exhibited the lowest overall mean burn rate (1.00 kg/hr).
The catalytic stoves had the second lowest overall mean burn rate (1.17 kg/hr).
The add-on retrofits had the third highest overall mean burn rate (1.56 kg/hr).
The conventional stoves had the highest overall mean burn rate (1.76 kg/hr). The
overall mean burn rates were distinctive, with differences between technologies
greater than 0.17 kg/hr for each technology classification.
Table 3-13 presents a "Student's t" statistical comparison of the emission rate
(g/hr and g/kg) data sets for the four technologies (catalytic stoves, add-
on/retrofits, low-emission stoves, and conventional stoves). The "t" statistical
test is used to determine the probability of two data sets being statistically
alike. In Table 3-13 the data sets from each of the four technology groups were
individually compared with the data sets from each of the other technology groups
(total of six comparisons).
For the gram-per-hour emission rate comparison, the "t" test indicates a relatively
high confidence level (less than 20% probability that the data sets are alike) that
3-96
-------
Table 3-13
"STUDENT'S T" STATISTICAL EMISSION RATE COMPARISON
Sample Population Characteristics
Stove Technology
Catalytic Stoves
Add-on/Retrofits
Low-Emission Stoves
Conventional Stoves
Grams per Hour3/
Mean
16.4
17.9
13.4
20.1
Si
8.4
9.3
10.2
9.5
Range
1.7-41.3
6.3-37.3
2.0-47.6
2.9-34.0
Nd/
58
19
23
14
Grams per Kilogram"'
Mean
14.4
13.2
13.2
12.1
Si
7.0
9.6
8.6
5.6
Range
1.2-34.1
5.3-37.9
1.4-37.9
1.2-22.0
Nd/
58
19
23
14
"Student's t" Comparison6/
Technologies
Compared
Catalytic Stoves
Add-on/Retrofits
Catalytic Stoves
Low-Emission Stoves
Catalytic Stoves
Conventional Stoves
Add-on/Retrofits
Low-Emission Stoves
Add-on/Retrofits
Conventional Stoves
Low-Emission Stoves
Conventional Stoves
Grams per Hour9'
rf/
0.64
1.35
1.40
1.46
0.65
1.93
DF9/
75
79
70
40
31
35
Ph/
50%-60%
10%-20%
10%-20%
10%-20%
50%-60%
<10%
Grams per Kilogram^/
rf/
0.57
0.62
1.08
0.01
0.35
0.40
DF9/
75
79
70
40
31
35
Ph/
50%-60%
50%-60%
20%-30%
>90%
70%-80%
60%-70%
a/ Particulate Emissions (g/hr)--Particulate emission rate in grams material per
hour of stove operation.
"/ Particulate Emissions (g/kg)--Particulate emission rate in grams material per
kilogram fuel (normalized to 0% moisture).
c/ ^--Standard deviation.
d/ N--Sample population.
e/ "Student's t" Comparison--A statistical comparison method used to determine
the probability of the means of two sample populations being alike; based on
range, standard deviation, and number of values for each sample population.
'' r--Air indication of the similarities between two sample populations. High T
values indicate significant differences between two sample populations; T values
near 0 indicate similarities between two sample populations.
9/ DF—Degrees of Freedom—Total number of values in two combined sample
populations minus two (i.e., Ni+N2-2). Degrees of freedom value is used in
conjunction with r value to determine the probability of the means of two sample
populations being alike.
h/ p--probability of the means of two sample populations being alike.
3-97
-------
the overall mean emission rate for the low-emission stoves (13.4 g/hr) is less than
the overall mean emission rate for the catalytic stoves (16.4 g/hr), add-
on/retrofits (17.9 g/hr), and conventional stoves (20.1 g/hr). There is also a
relatively high confidence level (10% to 20% probability that the data sets are
alike) that the overall mean emission rate for the catalytic stoves (16.4 g/hr) is
less than the overall mean emission rate for the conventional stoves (20.1 g/hr).
The probabilities of the emission rates being alike for the remaining two
comparisons (catalytic stoves vs. add-on/retrofits and add-on/retrofits vs.
conventional stoves) are inconclusive (50% to 60% for catalytic stoves vs. add-
on/retrofits, 50% to 60% for add-on/retrofits vs. conventional stoves). Of all the
stove group comparisons, only the mean emission rate of the low-emission stove
group is considered statistically different from the mean emission rate of the
conventional stove group.
For the gram-per-kilogram emission rate comparison, the "t" test shows a very high
confidence level (>90% probability of the data sets being alike) that the overall
mean emission rates for the low-emission stoves (13.2 g/kg) and add-on/ retrofits
(13.2 g/kg) are alike. There is a lower confidence level (20% to 30% probability
that the data sets are alike) that the overall mean emission rate for the catalytic
stoves (14.4 g/kg) is greater than the overall mean emission rate for the
conventional stoves (12.1 g/kg). There is a similar confidence level (70% to 80%
probability that the data sets are alike) that the data sets for the add-on/
retrofits (13.2 g/kg) and the conventional stoves (12.1 g/kg) are alike. The three
remaining emission rate comparisons (catalytic stoves vs. add-on/retrofits,
catalytic stoves vs. low-emission stoves, and low-emission stoves vs. conventional
stoves) have probabilities of the data sets being alike that are fairly uncertain
(50% to 60% probability for catalytic stoves vs. add-on retrofits, 50% to 60%
probability for catalytic stoves vs. low-emission stoves, and 60% to 70%
probability for low-emission stoves vs. conventional stoves).
Figure 3-11 summarizes the mean performance characteristics (particulate emissions,
wood use, and creosote accumulation) by stove type. The figure shows that the
relative ranking by stove technology is the same for all three parameters. For
each parameter presented, the low-emission stoves have the lowest values, followed
consecutively by the catalytic stoves, add-on/retrofits, and conventional stoves.
3-98
-------
Figure 3-11
Performance Comparison by Stove Technology
CO
10
30,0 -
25,0 -
HEftN
PflRTICULflTE
EHISSIDNS
(G/HR)
15.0 -
10.0 -
5.0 -
fiDD-DN/
RETROFITS
CfHRLYTIC
STDUES
LOU EMISSION
STOUES
•
CDNUEHTIDNflL
STD'UES -T-
PflRTICULflTE EMISSIONS (G/HR)
HOOD USE (KG/HDD)
CREOSOTE flCCUflULflTIDH (KG/1000 HDD)
• 1 SD
HEftN
-1 SD
- 2.00
- 1.80
- 1.50
HEflN
USE
(KG/HDD)
- 1.20
HEflN
CREOSOTE
ftCCUHULflTIQN
(KG/1000 HDD)
-0.50
- 0.20
-------
CATALYST EFFECTIVENESS
Introduction
Data on catalyst operation times are presented in Tables 3-10A and 3-11A. Table
3-12 presents additional detailed catalyst operational data.
Data are presented for each sampling period. Parameters include stove operational
time (%), catalyst operational time (%), average catalyst temperature (°C), average
catalyst AT (°C), percent of stove operational time that the bypass damper was open
(%), average flue gas temperature (°C), and particulate emission rate (g/hr).
The catalyst operational time is based on the percentage of time that the catalyst
was active (in-catalyst or catalyst-outlet temperature > 260°C [500°F]) while the
stove was active (flue gas temperature > 38°C [100°F]). The AT approach is another
technique that may be used to determine catalyst operational time. By this
definition, the catalyst is defined as being active when the in-catalyst or after-
catalyst temperature is greater than the before-catalyst temperature. However,
placement of the thermocouples and stove design factors are critical. If the
catalyst thermocouple could not be placed directly in the catalyst substrate (as in
the case of movable add-on catalysts), it was not possible to measure a true in-
catalyst temperature. In some catalytic stoves a temperature increase across the
combustor may not be indicated due to high pre-catalyst flue gas temperatures. As
indicated by the data, several negative AT values were measured. In another stove,
a positive AT could occur even with the stove bypass open. For these reasons, the
AT approach was not used to calculate catalyst operational time.
The percentage of stove operational time (flue gas temperature > 38°C [100°F]) that
the bypass damper was open is defined as the percentage of stove operational time
in which the flue temperature was greater than or equal to the catalyst
temperature. Thermocouple readings are recorded by the Data LOG'r once per minute
and these readings are averaged over 15-minute intervals, so only prolonged periods
of bypass use may be indicated. The bypass use percentages should therefore be
used with caution.
Combustor Replacement
Catalysts were changed in catalytic Stove Model D (four homes) between the 1985-86
and 1986-87 heating seasons. The catalysts used in the 1985-86 heating season were
composed of a cordierite-based ceramic, while the new combustors used in the 1986-
87 heating season were made of a mul1ite-based ceramic. Table 3-14 summarizes the
3-100
-------
Table 3-14
EFFECTS OF COMBUSTOR CHANGE ON PARTICIPATE EMISSIONS, BURN RATE, AND CATALYST OPERATION
STOVE CODE D
co
i—>
o
Home
Code
V08
V13
N02
Nil
ALL
Heating
Season
(Combustor)
85/86
(Codierite)
86/87
(Mullite)
85/86
(Codierite)
86/87
(Mullite)
85/86
(Codierite)
86/87
(Mullite)
85/86
(Codierite)
86/87
(Mullite)
85/86
(Codierite)
86/87
(Mullite)
Particulate Emissions
(g/hr)3/
Mean
19.3
12.0
18.6
11. 3f/
8.5
8.4
14.9
5.5
15.4
10.3
,"/
1.1
2.6
0.8
1.2
1.5
1.3
-0-
-0-
4.7
2.7
N.e/
2
4
2
4
2
3
1
1
7
12
Range
18.2-20.4
7.6-14.1
17.8-19.4
9.7-12.4
7.0—9.9
6.9-10.0
-0-
-0-
7.0-20.4
5.5-14.1
Burn Rate
(kg/hr)b/
Mean
1.15
0.89
0.92
1.20
1.11
0.99
0.90
0.88
1.05
1.00
,d/
0.14
0.20
0.19
0.02
0.06
0.14
-0-
0.24
0.17
0.22
Ne/
3
4
2
4
2
3
1
3
8
14
Range
0.96-1.27
0.61-1.09
0.73-1.10
1.18-1.24
1.05-1.17
0.82-1.17
-0-
0.90-1.16
0.73-1.27
0.61-1.24
Catalyst Operation
(Percent)0/
Mean
52.0
65.0
44.6
45.9
78.6
58.2
31.7
51.6
54.3
55.2
„"/
7.8
5.0
12.8
3.9
4.5
3.0
-0-
11.8
17.5
10.0
Ne/
3
4
2
4
2
3
1
3
8
14
Range
40.9-57.7
58.4-70.9
31.8-57.4
39.9-49.6
74.1-83.1
53.9-60.7
-0-
36.2-65.0
31.7-83.1
36.2-70.9
(Continued)
-------
o
PO
Table 3-14 (Continued)
EFFECTS OF COMBUSTOR CHANGE ON PARTICIPATE EMISSIONS, BURN RATE, AND CATALYST OPERATION
STOVE CODE D
a' Particulate emission rate in grains material per hour of stove operation (flue gas temperature >38°
C).
"' Average fuel consumption (normalized to 0% moisture) per hour of stove operation (flue gas
temperature >38° C).
c/ Percent of time the catalyst was in operation (>260° C) while the stove was in operation (flue gas
temperature >38° C).
d/ Standard deviation.
e/ Sample population.
'' Averaged results from two AWES sampling simultaneously.
-------
emission rate (g/hr), burn rate (kg/hr), and catalyst operation (%) data obtained
for each of the two heating seasons.
In all four homes the emission rate was lower in the 1986-87 heating season (using
mullite-based catalysts) relative to the 1985-86 heating season (using cordierite-
based catalysts). On an overall basis, the Stove D homes using the cordierite-
based catalyst had a mean emission rate of 15.4 g/hr, while the same stoves using
the mullite-based catalyst had a mean emission rate of 10.3 g/hr.
On an overall basis, the Stove D homes using the cordierite-based catalyst had a
mean burn rate of 1.05 kg/hr, while the same stoves using the mullite-based
catalyst had a mean burn rate of 1.00 kg/hr. The overall mean burn rate difference
between the two heating seasons (0.05 kg/hr) is probably not significant due to the
expected precision of the calculation method. Three of the four homes using Stove
Model D had decreased mean burn rates during the 1986-87 heating season (mullite-
based catalyst).
Three of the four homes using Stove Model D had increases in the overall mean
catalyst operational time percentage during the 1986-87 heating season (mullite-
based catalyst). On an overall basis, the stoves using the cordierite-based
catalyst had a mean catalyst operational time percentage of 54.3%, while the stoves
using the mullite-based catalyst had a mean catalyst operational time percentage of
55.2%. The difference in mean catalyst operational time percentage is not
statistically significant due to potential errors associated with the catalyst
operational time calculations.
The primary effect of the catalyst change between heating seasons appears to be a
reduction in emission rates. On an overall basis, the emission rate decreased by
34%. Changes in the overall mean burn rate and catalyst operational time
percentage appear to be insignificant. This analysis is remarkable for an in-situ
study in that several factors were presumably held constant between the two heating
seasons (stove installation characteristics, chimney systems, homeowner operating
practices, burn rate, and catalyst operational time), and emissions were measurably
reduced. As the Stove D units were experiencing combustor degradation (see the
following section, "Catalyst Longevity"), the emission rate reduction may simply
reflect "normal" performance by the stove when the combustors remained intact. The
mullite-based combustors may also have had other emission-reduction benefits.
3-103
-------
CATALYST LONGEVITY
Homes Using Existing Catalytic Stoves
Six homes in the NCS study had been using integrated catalytic stoves for
approximately one heating season prior to the study. These six homes (designated
Group III) were intended to provide a source of information on catalyst longevity,
as combustors in these homes had an additional year of use compared to the stoves
installed for the study. The existing catalytic stoves included six different
stove models with catalytic combustors from three different manufacturers (one
manufacturer's catalyst was used in four of the stove models). These six stove
models represented a range of catalytic technology.
At the end of the third heating season of stove use (May 1987), one of these six
existing catalytic stoves (V31) appeared to be working well using the original
catalytic combustor provided with the stove at purchase. Relatively good combustor
performance was evidenced by elevated temperatures downstream of the combustor
during the third year's use, by field emission results, and documentation of
creosote accumulation.
Three of the existing catalytic stoves had a combustor replaced during the study
due to inactivity or deterioration of the catalyst. One of the existing catalytic
stoves was found to have a deteriorated catalyst in the end-of-study stove
inspections, and the condition of the combustor in another existing stove was
unknown and suspected to be inactive. Home V33 was found to have erosion of the
combustor substrate material during the second year of use (1985/86 heating season)
and the combustor was replaced under warranty. Home N32 was found to have
experienced erosion of the combustor substrate material sometime during the third
season of use. Home V32 had an apparently inactive catalytic combustor (evident
from low temperatures) during its second heating season (1985/86) and was also
replaced under warranty. Home N33 experienced peeling of the catalyst coating of
the combustor beginning sometime in the first or second year of stove operation.
This peeling was accompanied by a loss of catalyst activity and the combustor was
replaced in the spring of 1987 (all emission tests were conducted at Home N33
before the combustor replacement). The condition of the catalytic combustor in a
fourth stove (N31) is unknown, but the homeowner suspects this metal catalyst has
become inactive. This combustor was not replaced during the course of the study.
Replacements are discussed at further length in the "Combustor Failures" section of
this report.
3-104
-------
Laboratory Testing of Field Combustors
During the first heating season, AWES testing results showed higher emissions from
catalytic stoves than might have been expected based on laboratory emission testing
of the same stoves and combustors. A testing program was implemented to address
potential causes by documenting combustor performance in standardized laboratory
tests after one year of field use.
The laboratory retesting of field combustors consisted of removing the catalytic
combustors from three study stoves after they had been in use for one (or two)
heating season(s) and installing the combustors into a laboratory control stove for
standard laboratory woodstove emissions testing. Combustors selected for retesting
were 14.5 cm (5.7 inch) diameter, 7.6 cm (3.0 inch) thick ceramic units from study
stoves with 1985/86 field results representing high (N03), medium (V07), and
relatively low (N32) emission results. (Home N32, as a Group III home, actually
had two seasons of use prior to retesting.) These combustors represented units
from two manufacturers.
The combustors were laboratory tested in a Blaze King Princess prototype stove,
which is the control stove used for combustor equivalency testing (for all
combustor models) under the State of Oregon Department of Environmental Quality
(DEQ) woodstove emissions certification program. Testing was conducted in
accordance with the Oregon DEQ "Standard Method for Measuring the Emissions and
Efficiency of Woodstoves, June 8, 1984" at the OMNI Testing Laboratory in
Beaverton, Oregon. An Oregon Method 7 (OM7) sampling system was used to measure
particulate emission rates. A single test was conducted at a burn rate of about
1.2 dry kg/hr, which is typical for the Northeast region. The laboratory retesting
of these field combustors provided a direct comparison of new combustor vs. used
combustor performance using the same control woodstove, the same testing method,
and the same laboratory for all testing.
Results of these tests are presented in Table 3-15 and Figures 3-12A through 3-12C.
The lowest laboratory results on a "field aged" combustor were from Combustor A
(Home N32, Stove P) with 6.5 grams per hour at a burn rate near 1.2 kg/hour. This
combustor was from a home which used the catalytic stove as a primary heat source
(typically 24 hrs/day during winter months) for two full heating seasons prior to
the laboratory tests. This combustor is estimated to have had about 6000 hours of
use at the time of retesting.
3-10-5
-------
Table 3-15
LABORATORY TEST RESULTS: NEW VS. USED COMBUSTORS3/
Combustor
Model
A
B
Status/Source
NEW (aged 50 hours prior to
testing)
USED (N32, used approximately
6000 hours in field)
NEW (aged 50 hours prior to
testing)
USED (N03, used approximately
1500 hours in field)
USED (V07, used approximately
3000 hours in field)
Lab Test Results
Burn Rate
(dry kg/hr)
0.80
0.98
1.20
1.50
1.20
0.66
0.80
1.60
2.09
1.26
1.17
1.23
Particulate Emissions
(g/hr)
0.8
0.9
1.7
2.2
6.5
1.7
0.8
2.7
6.9
7.41
30.4
21.5
a' All testing was conducted under laboratory conditions using certification
procedures (Oregon DEQ "Standard Method for Measuring the Emissions and
Efficiency of Woodstoves, June 8, 1984"). Particulate emissions were measured
using an Oregon Method 7 (OM7) sampling train. All combustors were tested while
installed in a Blaze King "Princess" prototype stove.
3-106
-------
Participate
Emissions
(g/hr)
CO
I—»
o
Figure 3-12A
Catalyst Longevity - Home N32, Stove P, Combustor A
D
o
8.2
8.6
1.8 1.4
Burn Rate (dry kg/hr)
0 Lab Test, New (OM-7)
o Lab Test, After Approximately 6889 Hours (Qfl-7)
D Field Test (AWES)
1.8
-------
Particulate
Emissions
(g/hr)
CO
I
Q
CO
Figure 3-12B
Catalyst Longevity - Home N03, Stove Cj Combustor B
D
D
0.2
6.6
1.0 1.4
Burn Rate (dry kg/hr)
0 Lab Test; New (Ofl-7)
o Lab Test, After Approximately 1500 Hours (Ott-7)
D Field Test (AWES)
1.8
-------
Figure 3-12C
Catalyst Longevity - Home v"B7j Stove Cj Combustor B
Particulate
Emissions
(g/hr)
o
vo
o
o
D
D
8.2
6.6
1.8 1.4
Burn Rate (dry kg/hr)
0 Lab Test, New (Ofl-7)
o Lab Test, After Approximately 3888 Hours (Ofl-7)
D Field Test (AWES)
D
1.8
-------
Laboratory results from Combustor B were obtained for two study homes, N03 and V07,
both of which used Stove C units. Home N03 used the catalytic stove as a back-up
to solar heating, with an estimated 1500 hours of use (one heating season) when the
laboratory testing was conducted. The laboratory results of the N03 combustor were
7.4 grams per hour at a burn rate near 1.3 kg/hour. The combustor from Home V07
was estimated to have about 3000 hours catalytic activity at the time of laboratory
testing (one heating season with full time use) and yielded laboratory test results
of 21.5 grams per hour and 30.4 grams per hour at burn rates near 1.2 kg. dry fuel
per hour.
All combustors were tested and reinstalled in their respective study home before
the start of the 1986-87 heating season.
Combustor A (Home N32, Stove P) results from lab retesting showed relatively low
emissions in comparison to original testing on a new combustor (Figure 3-12A).
With approximately 6000 hours of use, this combustor still showed the ability to
significantly reduce particulate emissions under laboratory test conditions. Field
testing with the AWES system showed one sample at over 20 grams per hour- The low
lab retest value indicates that the combustor, while showing some loss of
effectiveness with use, was still capable of significant particulate emission
reductions. The higher field sample was taken during the second heating season,
and may either indicate combustor degradation, stove maintenance needs, or fuel/
operator factors.
Combustor B (Home N03, Stove C) also showed relatively low emissions in comparison
to lab testing on a new combustor (Figure 3-12B). Field testing results with the
AWES sampler showed one sample to be quite similar to lab retest results, but that
other samples were significantly higher. All AWES samples were taken after the lab
retest. Again, this indicates that with 1500 hours of use, Combustor B was capable
of performing relatively well, but did not always do so in the test home. (It
should be noted that the combustor was installed in Stove C in the field, but lab
tested using the "Princess" prototype.)
The second combustor B unit (Home V07, Stove C) showed seemingly contradictory
results (Figure 3-12C). Lab retesting showed relatively high emissions. After a
retest result of about 30 g/hr, a second run was conducted with special care given
to maintaining catalyst activity during the start-up period. The second retest
showed emissions of about 22 g/hr. However, field testing before and after the lab
retest showed results ranging from 1.7 g/hr to 11.4 g/hr. The 1.7 g/hr sample may
3-110
-------
be considered an outlier due to the extremely low value. Field test burn rates
ranged from 1.4 to 1.9 kg/hr, compared to about 1.2 kg/hr for lab retesting. The
stove or operation factors are probable explanations for the relatively low field
values. The indication is that the stove was operated in the field in such a way
that emission's were consistently lower than when tested in the standard catalytic
stove using stove certification fuel loads and procedures.
Inspections
Between the 1985-86 and 1986-87 heating seasons, inspections of all catalytic
combustors were conducted to document the appearance of combustors and any problems
reported by homeowners. This inspection program was added to the study in Spring
1986 after several combustors were reported to be experiencing physical
deterioration. Due to the focus on catalytic devices at this point in the study,
only stoves or add-ons with combustors were inspected. An additional inspection of
all available study stoves was conducted before the start of the 1987-88 heating
season. All combustor problems noted in these September 1987 inspections are noted
here. Detailed results from this inspection will be available in a future report.
Thirty-five combustors were inspected at the end of the 1985/86 heating season; six
were in existing (Group III) stoves, while 29 were in new stoves or add-
on/retrofits. Seventeen were in integrated catalytic stoves, four were in
retrofits, and eight were in add-ons. Table 3-16 summarizes results of the
inspections. An assessment of observed combustor conditions is listed below.
Plugging. Ash and carbon build-up on the inlet face of combustors was the most
common combustor problem reported by inspectors, with 10 cases described. Ash
build-up ranged from minor (light build-up on a few cells) to severe, in some cases
occluding the combustor cells and causing smoke back-up and spillage into the room.
Heavy plugging was reported primarily in add-on devices. Five integrated catalysts
experienced some degree of ash build-up, with all but one reported as moderate or
minor. Five of the eight add-on devices experienced ash build-up, with three of
the five experiencing heavy build-up.
Cracking. Fracturing of a combustor substrate was found in only one case (N04).
It is not clear whether the combustor was broken during installation or during
operation of the unit. The combustor was replaced within one month after damage
was discovered. After a second combustor was installed, no further cracking was
noted.
3-111
-------
Table 3-16
1985-1986 HEATING SEASON COMBUSTOR INSPECTIONS
Technology
Group
Cat
Cat
Cat
Cat
Retro
Retro
Add-on
Stove
Model
A
B
C
D
E
F
G
1985-86
Combustor
Model
C
B
B
A
C
A
A & C
1985-86
Home
Code
N01
N10
V05
Vll
N09
N23
V07
V16
V17
V26
N03
N19
V08
V13
V18
N02
Nil
V01
V03
V12
N05
V02
N04
Combustor/
Substrate
Problem
minor peeling
none
none—stove installed
with combustor
improperly seated
none
none
none
none
none
none
none
none
none
substrate failure,
combustor replaced
substrate failure,
combustor replaced,
erosion showing on
2nd combustor
substrate failure
substrate failure
none
substrate failure
none
none
none
none
cracked combustor
"Ash
Plugging"
none
none
none
none
none
none
minor
none
none
none
none
none
heavy
none
none
none
none
none
none
none
none
none
none
Homeowner
Aware of
Problem
n/a
n/a
no
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
no
no
no
no
n/a
no
n/a
n/a
n/a
n/a
yes
(Continued)
3-112
-------
Table 3-16 (Continued)
1985-1986 Heating Season Combustor Inspections
Technology
Group
Add-on
Add-on
Existing
Catalytic
Stove
Model
H
I
P
1985-86
Combustor
Model
D
C
A
A or B
A
E
A
A
1985-86
Home
Code
V10
V15
N13
N06
N12
N14
V31
V32
V33
N31
N32
N33
Combustor/
Substrate
Problem
none
none
none
none
none
none
none «
"inactive" combustor
substrate failure
none
none
heavy peeling
"Ash
Plugging"
heavy,
repeated
heavy
none
moderate
minor
heavy
none
none
none
minor
minor
moderate
Homeowner
Aware of
Problem
yes
no
n/a
no
no
yes
no
yes
no
n/a
n/a
no
3-113
-------
Structural Damage. This category includes any type of severe deterioration of the
combustor structure. In virtually all cases of structural damage, crumbling and
extensive erosion occurred on the inlet face, resulting in a hole forming through
the combustor or parts falling out of the sheath. Five of the failures were in the
catalytic stoves, with one repeated case in a retrofit design. One additional
structural failure in a catalytic stove was discovered in the September 1987 stove
inspection. A majority (four of seven) of the failed combustors were in a single
stove design (Stove D), five of seven were the same combustor model, and six of the
seven were the same combustor manufacturer. Combustors using a new mullite-based
substrate material were installed in all Stove D units before the start of the
1986-87 heating season (see Catalyst Effectiveness section).
The homes experiencing combustor substrate deterioration included six homes using
catalytic stoves and one home with a retrofit-equipped woodstove. It is notable
that six of the seven deteriorated combustors were the same brand. It is also
important to note that four of the seven failures were of one stove design type,
all using the same type of catalytic combustor. All of the stoves of this model
used replacement "second generation" combustors from the same catalyst manufacturer
during the second heating season. Two of these "second generation" catalysts
exhibited minor evidence of surface erosion noted on previous combustors. One of
these combustors was in a stove with a poorly-sealed ash pan door, which probably
caused frequent over-firing.
The seven original combustors were observed experiencing deterioration of the
substrate material. In the earliest stage of observed damage, the upstream surface
(bottom) of the combustor was eroded near the center of the combustor. As this
deteriorating continued, the erosion became deeper and affected a larger frontal
area of the combustor, decreasing the effective thickness of the combustor.
Finally, the continuing erosion of the substrate resulted in a hole, which
continued to increase in size until eventually the entire combustor physically
crumbled. During this deterioration, the remaining portion of the catalytic
combustor appeared active, as evidenced by a glowing combustor with elevated
temperatures. In at least one case the homeowner did not notice catalyst
deterioration from a decrease in temperatures on the supplied catalyst monitoring
thermometer, as it was located above a section of the catalyst not affected by
deterioration.
Combustor substrate deterioration can increase particulate emissions significantly,
especially if a hole develops in the combustor. Samples N02-1, -2, and -3, and
3-114
-------
V01-4, -5, -6, and -7 in Table 3-10C offer examples. Actual inspections of the
combustor were not made immediately before or after any of these samples. Home N02
had a crumbled catalyst replaced between samples N02-2 and N02-3. It is assumed
that some erosion was occurring during sample N02-1 and that severe erosion was
likely during sample N02-2. It is likely the erosion during the first sample (N02-
1) was small and all the exhaust gases were passing through a combustor cell,
although the catalyst may have been eroded to less than a two-inch thickness across
several cells. The second sample (N02-2), exhibiting a marked increase in
particulate emissions, was collected shortly before a severely deteriorated
catalyst was removed from the stove. The third sample (N02-3) was taken after the
combustor had been replaced, and shows significantly reduced emissions.
Home V01 experienced two combustor substrate failures resulting in crumbled
catalysts. The September 1987 combustor inspections also revealed the third
combustor in use at this home was brittle, and crumbled during removal despite
gentle handling. One of the failed combustors was replaced between samples V01-6
and V01-7. Sample V01-6 shows the highest particulate emissions obtained with this
retrofit catalytic stove. Sample V01-7, collected just after the catalyst was
replaced, shows significantly reduced emissions.
Home V13 had a crumbled catalyst replaced just before sample V13-2. However, the
replacement catalyst was also found deteriorated at the end of the 1985/86 heating
season. As a result, samples V13-2 and V13-3 were also collected during a period
of combustor erosion. Samples V05-1 and V05-2 did not have a damaged combustor
substrate, but did have the combustor out of position.
Peeling. Some combustors have a "wash coat" to provide higher surface area for
catalytic activity. Peeling of the wash coat was observed in two cases (N01 and
N33) on combustors in catalytic stoves. No peeling was noted in add-on or retrofit
devices. The homeowners were not aware of the combustor peeling, but indicated
they would replace the combustor if they thought it was inoperative.
Erosion. Erosion appears as crumbling of portions of the combustor. It usually
begins in the center of the inlet face and appears to grow progressively wider and
deeper. Though apparently related to most cases of structural failure, it is not
obvious what variables directly cause erosion. The amount of deterioration was
variable between stoves. Direct flame impingement and excessive temperatures are
thought to be contributing factors. It should be noted that on one stove (D model,
Home V13), erosion was appearing on the replacement (cordierite) combustor when
3-115
-------
end-of-season inspections were conducted. Erosion occurred on the same model of
combustor in all but two cases and the same brand in all but one case. Subsequent
replacement with mullite-based combustors during the 1986-87 heating season
resulted in no evidence of erosion on the Stove D combustor in V08, but slight
erosion on the V13 combustor. (Note that V13 was the installation with the poorly-
sealed ash pan door.)
Other. One combustor (Home V05) was found to be ineffective due to improper
positioning in the combustor support "cup." The combustor was either not installed
properly or was jarred out of position during transport and installation. Another
combustor (V32) was noted by the homeowner to attain lower temperatures in contrast
to the previous heating season, and failed to "glow" (this combustor had one year
of previous use). Particulate emission samples collected from this home show a
marked decrease in emissions after the replacement of the "inactive" combustor. In
several other instances, gaskets used to assure a tight seal around the combustor
appeared to be deteriorating. A second case of an apparently inactive combustor
was discovered in Home V16 from temperature data collected by the Data LOG'r system
as well as by the change in color on a stove bypass rod. This stove design
features a catalytic bypass rod assembly which passes near the downstream surface
of the catalytic combustor. The color of this bypass rod when pulled out can
indicate combustor performance; a light gray soot indicating proper performance and
a dark brown or black color usually indicates improper catalytic operation or an
inactive catalyst. The definite change in creosote color on this bypass rod during
the second heating season of the study under the same operating conditions appears
to indicate the combustor was inactive. The combustor in this stove did not show
elevated catalyst temperatures during the 1986/87 data collection period. In this
case, the homeowners at study Home V16 were not aware of the catalyst failure until
coloration of the bypass rod was pointed out to them (even though the
manufacturer's instruction manual mentioned coloration as an indication of poor
catalyst performance).
Most of the damaged combustors and the poorly seated combustor were discovered as
incidental observations to chimney sweeping tasks or by combustor inspections by
field test personnel. Homeowners reported they did not notice any change in stove
performance throughout the heating season. Four cases of combustor substrate
crumbling were noted by the chimney sweeps during study mid-season chimney
cleanings (Homes V13, V18, V33, and N02). In each of these cases, the catalyst
component was readily visible to the sweep without dismantling the appliance
itself; hence their discovery. Two other cases of substrate crumbling occurred in
3-116
-------
a retrofit stove (Home V01 in both cases) where the catalyst component was not
readily visible to the sweep; the damaged catalyst was discovered by field test
personnel, who noted that pieces of the combustor had fallen through the flame
baffles into the firebox. An additional case of substrate crumbling at Home V13, a
recurrence of catalyst deterioration in a stove where the original catalyst had
been replaced, and a case of catalyst peeling (Home N33) were discovered during
mid-study catalyst inspections. An additional case of substrate crumbling at Home
N32 was discovered during the end-of-study catalyst inspections. The poorly seated
combustor was discovered by field personnel looking for causes of the heavy chimney
creosote accumulations at this home (V05). (A chimney fire eventually occurred in
this home.) Minor dismantling of the stove was required to access the combustor.
Four of the cases listed above occurred in the Group III homes which were using
their own catalytic stoves before the start of the study. One case of substrate
crumbling (V33 and N32), one case of catalyst peeling (N33), and the case of an
apparently inactive catalytic combustor (V32) were noted. It is possible that some
of the combustor problems in the three existing stoves may have started during the
first year (1984-85) of heating. The remainder of cases were in appliances
provided by the study and being used in their first or second year of heating.
Combustor Replacements
Over the course of the two-year study period, a total of 13 combustors were
replaced. Nine catalysts were replaced due to a failure of the previous unit.
Four catalysts were replaced between the 1985-86 and 1986-87 heating seasons to
permit testing of a mullite-based combustor. The mullite-based replacements were
all installed in the same stove model which had four cases of ceramic substrate
deterioration (using the original combustors) during the first heating season
(1985-86). The original and replacement combustors were supplied by the same
catalyst manufacturer. The nine combustors replaced due to failures were replaced
with the same type combustor as previously used. Table 3-17 lists combustor
replacements.
3-117
-------
Table 3-17
COMBUSTOR REPLACEMENT CHRONOLOGY
Study Home
V13
N04
V33
N02
V18
V01
V32
V08
V13
N02
Nil
V01
N33
Stove
Code
D
G
P
D
D
E
P
D
D
D
D
E
P
Time of Replacement
(Month/Year)
2/86
2/86
3/86
3/86
3/86
4/86
10/86
10/86
10/86
10/86
10/86
2/87
3/87
Reason
Substrate deterioration
Cracked deterioration
Substrate deterioration
Substrate deterioration
Substrate deterioration
Substrate deterioration
Inactive
Switch to new type combustor
(substrate deterioration)
Switch to new type combustor
(substrate deterioration)
Switch to new type combustor
Switch to new type combustor
Substrate deterioration
Peel ing
Operator Factors
The combustor problems addressed above may be caused by several factors, including
stove operation. Interviews with homeowners were conducted in efforts to document
stove operation factors which might contribute to high (or low) emissions. (It
should also be noted that none of the low-emission or conventional stoves were
inspected for deterioration of gasketing or other factors.) Key issues identified
by the interviews are presented below:
1. Opening the ash door on some stove models can cause "underfire"
conditions where air enters the fuel load from below. This, or
leaving the fuel loading door open for extended periods, can cause
overfiring or flame impingement on a combustor. About half of
Vermont homeowners reported that they left either the ash door
(Stove D only) or fuel loading door open, but most said this was
only for starting the fire and was not done with the combustor
engaged. (The New York interviewer asked specifically about ash
door opening only. No homeowners reported opening an ash door, if
present, while the combustor was engaged.)
2. Catalyst poisoning was addressed by asking homeowners "if any trash,
plastic, or colored paper" was burned in the stove. Virtually all
3-118
-------
homeowners reported that only wood was burned. Three homeowners
stated that small amounts of cardboard or colored paper were burned,
but no catalyst deterioration was noted in these homes. Moderate to
heavy ash plugging was noted in the two add-on devices burning this
material.
3. Smoke spillage was used as an indicator of ash plugging problems or
poor draft conditions. About one third of catalyst owners reported
some backpuffing, although most reported problems only during start-
up or warm weather conditions.
4. Catalyst damage from overfiring was investigated by asking
homeowners if they operated the stove for long periods of time with
the thermostat or air control in a full open position. About one
fourth of catalyst owners reported that they did operate at high
fire conditions, although most said they did so only to establish
the fire. Three catalyst owners operated with the air supply fixed
open, but no apparent combustor damage was noted on these
combustors.
5. Homeowners were asked if they noted any change in stove or combustor
performance during the year. This question was intended to
determine if homeowners would be aware of a deteriorated or
malfunctioning combustor- Four homeowners reported a noticeable
change in combustor performance; three of these had confirmed or
suspected combustor failure. However, five cases of combustor
deterioration were not detected by homeowners.
6. Homeowners were also asked if they would replace a malfunctioning
combustor. Virtually all respondents said they would, although four
indicated that while they would replace it once, they would not do
so routinely. One homeowner said that he would switch from an add-
on to an integrated catalytic if the combustor failed.
Stove Design
Four of seven integrated or retrofitted catalytic stoves experiencing combustor
failure during the first heating season were in a single stove design (Stove D) and
combustor type. This stove model is relatively small (38 liters [1.4 ft^]); the
next-largest catalytic stove model provided to the study had a firebox size of 69
liters (2.5 ft^). The small size may have required homeowners to run the stove
near maximum levels to meet the heating demand of the home. A high burn intensity
(percent of fuel load burned per hour) can result in hot firebox conditions and
increase the likelihood of flame impingement on the combustor. Other combustor
failures occurred in stoves with firebox sizes of 52 liters (1.8 ft^) and 62 liters
(2.2 ft3). One combustor failure was in an existing catalytic stove model (firebox
size of 87 liters [3.1 ft^]) which had a damaged thermostatic control, preventing
the stove from being adjusted to lower heat outputs. Stove A models had cordierite
combustors, while Stove B had non-cordierite combustors.
3-119
-------
Underfire air or air entering the stove at low levels can cause rapid release of
the volatile compounds in the fuel load and possible overfiring of a combustor.
Stove D uses an underfire air configuration for primary air. In one home, the ash
pan access door was kept open to allow more primary air to enter under the fire.
In another case, gasket material around the ash pan access door deteriorated and
was not repaired.
One of the stoves (Home V01, Retrofit E) with deteriorated combustors did not have
direct underfire air, but had the potential for inducing a similar condition in the
firebox due to the location of the primary air supply. In this case, the primary
air inlet was approximately 7.5 cm (3 inches) above the bottom of the firebox. The
build-up of ash in the firebox raised the base of the fuel load until most of the
air supply was introduced below the fuel load. This condition can create the
potential for fuel-rich gas mixtures and high catalyst temperatures.
Flame baffles and secondary air inlets also appear to be significant stove design
factors in protecting the combustor from excessive temperatures. Many stove
designs use a combined secondary air inlet and "flame baffle" to mix fuel gases and
secondary air- In the stove design with the most combustor failures (Stove D), it
was speculated that a "secondary flame" from preignition (ignition of gases, due to
the addition of secondary air, before reaching the combustor) of fuel gases may
have routinely reached the combustor. This could have caused temperatures in
excess of 980°C (1800°F) and possible combustor damage. Incidentally, this could
also have resulted in negative AT readings across the combustor.
Combustor Factors
With one exception, all combustors experiencing severe erosion, crumbling, or
structural damage were from the same manufacturer. The manufacturer has stated
that this model is being replaced with a combustor constructed of non-cordierite
material which is less prone to deterioration. These combustors were placed in all
Stove D installations before the beginning of the 1986-87 heating season. No
substrate deterioration was noted in any of the Stove D units, although two
combustors did exhibit a chip out of the combustor face. The combustor
manufacturer states that this chip is characteristic of a thermal stress fracture.
Combustor temperature data from the first heating season do not appear to show
evidence of excessive temperatures. The maximum average operating temperature of a
catalytic stove combustor was 492°C (918°F) in Home N32. (The combustor in Home
N32 did experience structural breakdown in its third year of operation.) The
3-120
-------
maximum peak temperature for any stove combustor was 1020°C (1868°F). No visual
evidence of combustor deterioration was noted with this combustor.
POM AND TCO EMISSIONS
A subset of the particulate emission samples was selected for analysis for
polycyclic organic material (POM) and total chromatographable organic (TCO)
compounds. ROMs, as a group, have been demonstrated to have mutagenic properties.
Selected AWES samples were analyzed for eight ROMs by gas chromatography and mass
spectroscopy (GC/MS) and by gas chromatography with a flame ionization detector
(GC/FID) for TCOs. TCO compounds are defined as hydrocarbon compounds with boiling
points between 100°C and 300°C. ROMs and TCOs were measured in the combined
solution of the XAD-2 resin extract, filter catch extract, and solvent washings of
these samples. These analyses were intended to provide representative indications
of POM and TCO concentrations and emission rates from the stove technology groups.
Samples were analyzed for the following POM compounds: naphthalene, acenaphthene,
phenanthrene, pyrene, benzo(a)pyrene, indeno(l,2,3-c,d)pyrene, benzo(g,h, i)-
perylene, and 3-methyl cholanthrene. Some of these compounds, such as naphthalene,
are considered non-carcinogenic. (Naphthalene represented the largest mass
fraction of any of the POM compounds in all samples.) Others, such as
benzo(a)pyrene, have been demonstrated to be carcinogenic.
Several factors should be considered in evaluating POM and TCO results. Results
are presented in units of g/m^, g/hr, and as a fraction of total particulate mass
(Table 3-18A and 3-18B). The concentration (g/m^) and emission rate (g/hr) values
can be evaluated in terms of relative emission rates. However, the methods and
procedures used to collect, process, and analyze these samples necessarily limit
the accuracy of the reported values. Additionally, due to single samples in the
case of all stove technologies except catalytic, results should not be considered
representative. Reported values should be used only for order-of-magnitude type
evaluations.
Special care should be used in evaluating POM and TCO values from samples collected
before and after a catalyst. While samples were collected before and after the
combustor, only the flue collar sampler of the pair of AWES was equipped with an
oxygen cell for determining flue gas flow. All the catalytic stoves had secondary
air introduced between the firebox sampler and the flue collar sampler. As
secondary air flow was unquantified, some of the apparent POM and TCO reductions
may be due to dilution by secondary air. However, it is unlikely that secondary
3-121
-------
Table 3-18A
POM AND TCO EMISSIONS (g/m3)
oo
I
Stove
Technology
Catalytic
Retrofit
Add-On
Low-Emission
Conventional
Stove
Model
A
B
C
D
E
J
M
0
Sample
Code
N10-6
N10-6
Sampl ing
Location
Firebox
Flue Collar
Difference
Vll-7
Vll-7
Firebox
Flue Col lar
Difference
V07-6
V07-6
Firebox
Flue Collar
Difference
N02-6
V01-7
V10-7
V34-7
V06-6
Flue Collar
Flue Collar
Device Exit
Flue Collar
Flue Collar
POM Compounds3/0/ (g/m3)
A
0.03
0.01
0.02
0.02
0.01
0.01
0.02
0.01
0.01
0.03
0.01
f/
g/
g/
B
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
f/
g/
g/
c
g/
g/
g/
g/
g/
g/
g/
g/
g/
0.01
g/
f/
g/
g/
D
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
f/
g/
g/
E
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
g/
f/
g/
g/i/
F
g/
g/
g/
g/i/
g/i/
g/i/
g/
g/i/
g/
g/
g/
f/
g/
g/i/
G
g/
g/i/
g/i/
g/
g/i/
g/i/
g/
g/
g/
g/
g/
f/
g/
g/i/
H
g/
g/i/
g/i/
g/i/
g/i/
g/i/
g/i/
g/i/
g/i/
g/
g/
f/
g/i/
g/i/
Total J/
0.03
0.01
0.02
0.02
0.01
0.01
0.02
0.01
0.01
0.04
0.02
f/
0.01
g/
TCob/c/
(g/m3)
1.0
Od/
l.Qd/
0.2
0.1
0.1
0.2
e/
e/
0.1
0.4
f/
Od/
0.1
(Footnote references listed following Table 3-18B.)
-------
Table 3-18B
POM AND TCO EMISSIONS (g/hr)
Stove
Technology
Catalytic
Retrofit
Add-On
Low-Emission
Conventional
Stove
Model
A
B
C
D
E
J
M
0
Sample
Code
N10-6
N10-6
Sampling
Location
Firebox
Flue Collar
Difference
Vll-7
Vll-7
Firebox
Flue Collar
Difference
V07-6
V07-6
Firebox
Flue Collar
Difference
N02-6
V01-7
V10-7
V34-7
V06-6
Flue Collar
Flue Collar
Device Exit
Flue Collar
Flue Collar
POM Compounds3/0/ (g/hr)
A
0.5
0.2
0.3
0.5
0.2
0.3
0.5
0.3
0.2
0.4
0.3
f/
0.1
0.1
B
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
f/
h/
h/
C
h/
h/
h/
h/
h/
h/
0.1
0.1
0.0
0.1
h/
f/
h/
h/
D
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
f/
h/
h/
E
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
h/
f/
h/
h/i/
F
h/
h/
h/
h/i/
h/i/
h/i/
h/
h/i/
h/
h/
h/
f/
h/
h/i/
G
h/
h/i/
h/i/
h/
h/i/
h/i/
h/
h/
h/
h/
h/
f/
h/
h/i/
H
h/
h/i/
h/i/
h/i/
h/i/
h/i/
h/i/
h/i/
h/i/
h/
h/
f/
h/i/
h/i/
TotalJ/
0.6
0.2
0.4
0.5
0.2
0.3
0.6
0.5
0.1
0.6
0.4
f/
0.1
0.1
TCOb/c/
(g/hr)
19.7
od/
19. 7d/
5.6
2.5
3.1
6.4
e/
e/
1.4
8.4
f/
Qd/
3.4
(Footnote references listed following Table 3-18B.)
-------
f and 300° C.
ro
Table 3-18
POM AND TCO EMISSIONS
a' All catalytic devices had secondary air introduced between the fire box and the flue collar
samplers. Flow was calculated at the flue collar only. Reduction of concentration (g/nH) and
emission rates (g/hr) are therefore partly due to dilution of sample. Samples were analyzed for the
following Polycyclic Organic Materials (POMs):
A - Naphthalene E - Benzo(a)pyrene
B - Acenaphthene F - Ideno(l ,2,3-c,d)pyrene
C - Phenanthrene G - Benzo(g,h, i)perylene
D - Pyrene H - 3-Methyl Cholanthrene
Other POM compounds were undoubtedly present, but samples were not specifically analyzed for compounds
other than those listed above.
"' Total Chromatographable Organics, defined as hydrocarbon compounds with boiling points between 100°
c/ It should be noted that samples were collected over a one-week period and were not chilled or
refrigerated during this time, or during subsequent shipment to the laboratory. It is therefore
unclear how representative the samples may be of actual "real time" emissions.
d/ < field blank values.
e' TCO value > total particulate value.
f/ Low total hours of stove use — emission data questionable.
9/ < 0.00 g/m3
n/ < 0.0 g/hr
""/ Not detected.
J/ Total POM represents the sum of the eight compounds listed in note a/ above. Totals may not appear
to be a direct sum of values shown due to the cumulative total of values lower than the minimum
reporting values used here.
-------
air flow would equal the gas flow exiting the firebox; the three samples from
N10-6, Vll-7, and V07-6 probably showed some actual reduction in POM and TCO
concentrations.
Emissions may also be evaluated in terms of the relative POM and TCO mass fractions
of total particulate mass. This allows determination of whether various compounds
are selectively destroyed in a catalytic combustor, and is unaffected by secondary
air or flow issues. Table 3-19 lists results of three samples.
All samples showed reductions in total mass of the eight specified POM compounds.
While V07-6 showed effectively no change, N10-6 and Vll-7 showed probable
reductions of about a factor of two. Total TCO mass was reduced in N10-6 and Vll-7
samples, but showed an increase of about 370% in sample V07-6. This may be from
"catalytic cracking" of heavier organic compounds, as the total particulate mass
(while not directly comparable) is about three times lower after the combustor.
Reductions in the POM and TCO mass fractions of total particulate mass were
variable. Sample N10-6 showed a nominal reduction in the POM mass fraction, while
samples V07-6 and Vll-7 showed POM mass fraction increases of about 140% to 240%.
TCO mass fractions were apparently reduced to non-detectable levels in sample
N10-6, but showed increases of about 230% to 860% in samples Vll-7 and V07-6.
Due to the wide range of values and contradictory trends, it is difficult to draw
conclusions on catalyst efficiency at reducing POM and TCO fractions of the total
particulate mass. It appears that while the total mass of POM or TCO may be
reduced, they may not be reduced proportionately as a fraction of total particulate
mass.
3-125
-------
Table 3-19
POM AND TCO MASS FRACTIONS
Stove
Model
A
B
C
Sample
Code
N10-6
N10-6
Vll-7
Vll-7
V07-6
V07-6
Sampl ing
Location
Firebox
Flue
Collar
Firebox
Flue
Collar
Firebox
Flue
Collar
Total
Particulate
Mass (mg)
865.3
357.9
427.6
61.9
431.9
164.8
Total
POM Mass
(mg)
9.12
3.60
4.22
2.10
5.84
5.56
Total
TCO Mass
(mg)
294
0
45
22
59
222
POM
Fraction
of Part
(%)
1.1
1.0
1.0
3.4
1.4
3.4
TCO
Fraction
of Part
(%)
34
0
11
36
14
135
POM
Fractional
Reductions
Eff. (%)
9
-240
-143
TCO
Fractional
Reductions
Eff. (%)
100
-227
-864
-------
Section 4
ANALYSIS
INTRODUCTION
Although the data set of participate emissions, wood use, and creosote accumulation
is relatively large from this study, the wide range of data and the number of
variables make it difficult to identify the cause of data variability. As with any
case where clear and defensible conclusions are not easily identified, there is an
increased risk of misinterpretation of the data. In efforts to identify major
factors affecting stove performance, a more detailed examination of the data was
performed. This examined the various permutations of different elements or factors
which may affect stove performance.
Several factors must be considered in evaluating results from this study:
1. Objectives—this study was conceived as an evaluation of typical
stove operation under "real world" conditions. It was not intended
to demonstrate the best possible performance by woodstoves. It was
intentionally designed to emphasize catalytic technology; at the
time the project was conceived, low-emission non-catalytic
technology was largely still in the development stage.
The emphasis of the study was shifted at mid-point to provide more
balanced data on all low-emission technologies. Interest in
performance of the low-emission stoves was also increased with the
announcement of EPA's planned New Source Performance Standards
regulation of woodstoves. (Federal Register, Wednesday, February 18,
1987, pp. 4994-5066.)
2. Stove models—catalytic, catalytic add-on and retrofit, and low-
emission non-catalytic stoves were provided to the study by stove
manufacturers willing to participate in the project. Therefore,
stoves were not necessarily considered to be the best available,
although several stoves did show good performance in lab testing.
Most devices had been tested under laboratory conditions prior to
inclusion in the study.
Two low-emission non-catalytic stoves (M and N) and an add-on device
(J) were added to the study for the second heating season. These
units were selected based on lab testing results which showed them
to be among the best performers in their class. For this reason,
average performance by technology group may be biased in favor of
the low-emission group. No stove model changes were made in the
catalytic stove group, although combustors were replaced as needed.
4-1
-------
3. Climate—most laboratory testing to date has been conducted to meet
Oregon DEQ standards, which use "weather weighting" to emphasize
results at burn rates estimated to be typical of the mild Oregon
climate.
The colder Northeast region requires higher burn rates to meet the
heat demand. Lab tests show that most catalytic stoves have higher
emissions at higher burn rates, while most non-catalytic
(conventional and low-emission) stoves have lower emissions at high
burn rates. Comparisons to "weather-weighted" Oregon DEQ lab
results may therefore be inappropriate.
4. Test methods—the AWES system is considered to be essentially
equivalent to or have slightly higher results than the laboratory
methods of EPA 5G or 5H (OM7). Precision is also considered to be
essentially the same as the other methods (10-20%). While good
sampler precision allows comparison of AWES results, care should be
used when comparing results obtained with other methods due to
significant differences in fueling and operating practices.
BURN RATE EFFECTS ON PARTICIPATE EMISSIONS
Analysis of Data
The following analysis focuses primarily on the relationship of burn rate to
emission rate. It is recognized that other factors influence this relationship;
however, the purpose of this analysis is to identify whether a relationship of burn
rate to emission rate exists for a particular technology classification or stove
model. Other influences on emission rate (fueling patterns, chimney systems,
catalyst operation) will be discussed in later portions of this section.
Figures 4-1A through 4-1D and Figures 4-2A through 4-2D are plots of particulate
emissions (g/hr and g/kg) versus burn rate for each of the four woodstove
technologies evaluated (catalytic stoves, add-on/retrofits, low-emission stoves,
and conventional stoves). The figures use unique symbols for each stove model
(with the exception of the conventional stoves and existing catalytic stoves, which
were not separated by model). Linear regression coefficients (r2 values) were
calculated for each stove model with three or more valid data points and for the
total data set for each technology category. The r2 values indicate the closeness
of fit of the data points to a constant slope; an r2 value of 1.000 indicates that
the data points lie in a straight line, while an r2 value of 0.000 indicates no
apparent correlation, as in the case of three points forming an equilateral
triangle on the plot.
Catalytic Stoves. Figure 4-1A presents emission rate (g/hr) versus burn rate
(kg/hr) data from the catalytic stoves. The r2 values for the individual stove
4-2
-------
50 -
¥5 -
to -
PftRTICULftTE
EMISSIONS '
(G/HR)
30 -
25 -
20 -
15 -
10 -
5 -
Figure 4-1A
Particulate Emissions (g/hr) vs. Burn Rate (fcg/hr)
Catalytic Stoves
o
D
D
0.5
1.0
1.5
BURN RflTE (KG/HR)
flLL (R2 =0.133)
0 STDUE fl (R2 =0.118)
O STDUE B (R2 =0.3t8)
2.0
D STDUE C (R2 =0.223)
| STDUE D (R2 =0.051)
X STDUE P (R2 =0.811)
2.5
-------
50 H
to -
PfiRTICULflTE
EMISSIONS -
(G/HR)
SO -
25 -
20 -
15 -
10 -
r
Figure 4-iB
Particulate Emissions (g/hr) vs. Burn Rate Qg/hr)
Add-Qn/Retrofits
D
D
DY
D D
°
I
0
0.5
1.0
1.5
BURN RflTE (KG/HR)
2.0
2.5
ALL (R2 =0.005)
0 STOUE E CR2 =0.395)
O STQUE G (R2 =0 .832)
D STDUE F (R2 =0.273)
I STDUE J (R2 =0 .213)
X STDUE I CR2 =0.586)
Y STDUE H (R2 = Nft )
-------
50 -
to -
PflRTICULftTE
EMISSIONS *
(G/HR)
JO -
25 -
20 -
15 -
10 -
5 -
Figure 4-iC
Participate Emissions (g/hr) vs. Burn Rate (kg/hr)
Low-Emission Stoves
a
D
0
0.5
1.0
1.5
BURN RfiTE (KG/HR)
ftLL (R2 =0,095)
0 STDUE H (R2 =0.002)
O STaUE L (R2 =0 . 120)
1
2.0
D STDUE H (R2 =0.5S7)
| STDUE K (R2 =0.811)
2.5
-------
to -
PflRTICULflTE
EMISSIONS
(G/HR)
-pi
I
en
30 -
20 -
15 -
10 -
Figure 4-1D
Participate Emissions (g/hr) vs. Burn Rate Qg/hr)
Conventional Stoves
I
I
Is
i
I
0,5
i .0 1.5
BURN RftTE (K5/HR)
2,0
2.5
-------
Figure 4-2A
Participate Emissions (g/kg) vs. Burn Rate (fcg/hr)
Catalytic Stoves
50 -
fS -
to -
PflRTICULflTE
EHISSIDHS ~
o
• 00
• « : * '
' ' X*tf°o 0
i f? 5° .x
•° t *F ° n
nJ_ • n
• cP OQ D <7 D
D
0.5
1.0
1.5
BURN RflTE CKG/HR)
2.0
2.5
0 STDUE fl (R2 ro.
O STDUE B (R2 =0.OS9)
D STDUE C (R2 =0
| STDUE D (R2 =0.112)
X STDUE P tR2 =0.001)
-------
I
Co
50 -
t5 -
to -
PftRTICULATE
EMISSIONS '
(G/KG)
JO -
25 -
20 -
15 -
10 -
5 -
Figure 4-2B
Particulate Emissions (g/kg) vs. Burn Rate (kg/hr)
Add-On/ReirQfits
D
Y
0
10
v*
*
0.5
1.0
1.5
BURN RflTE (KG/HR)
2.0
2.5
ALL (R2 =0.2t2)
0 STOUE E (R2 =0 .213)
O STOUE G (R2 =0.695)
D STDUE F (R2 =0.550)
I STDUE J (R2 =0.t80)
X STDUE I CR2 =0.681)
Y STDUE H (R2 = Nfi )
-------
I
10
Figure 4-2C
Particulate Emissions (g/kg) vs. Burn Rate (kg/hr)
Low-Emission Stoves
50 -
ts -
pftRTICULftTE
EMISSIONS ~
(G/KG)
JO -
25 -
20 -
15 -
10 -
5 -
•
f
•
D
o
• ' o
CL°A
D GO. o
U o
o o
0
I 1 1 1 1
0.5 1.0 1.5 2.0 2.5
BURN RflTE (KG/HR)
ALL (R2 =0.012)
0 STDUE N (R2 =0.0f4)
O STDUE L (R2 =0.^62)
D STDUE H (R2 =0.t30)
| STDUE K (R2 =0.751)
-------
fO
PfiRTICULfiTE
EMISSIONS
(G/KG)
30
25 -
15 -
10 -
Figure 4-2D
Particulate Emissions (g/kg) vs. Burn Rate (kg/hr)
Conventional Stoves
R2 =0.i6f
0.5
i
§
i®
I
T
}
1.0 1.5 2.0
BURH RflTE (KS/HR)
I
I
2 .5
-------
models range from 0.051 (Stove D) to 0.348 (Stove B). Stove P, which includes
several models of catalytic stoves that were in place prior to the beginning of the
study, had the highest r2 value (0.811). The overall r2 value for catalytic stove
technology (0.133) indicates a poor emission rate/burn rate correlation.
If the outer data points on the plot are connected, a rough ellipse is formed which
has the long axis tilted to indicate an apparent trend of increased particulate
emission rates with increased burn rate. This apparent trend supports the general
hypothesis for the relationship of emission rate to burn rate for catalytic
technology; conventional wisdom states that increased emission rates (g/hr) are
expected with increased burn rates due to decreased flue gas residence time in the
catalyst.
The majority of data points in Figure 4-1A appear to be grouped in a rectangle
bounded by burn rates of 0.6 to 1.4 kg/hr and emission rates of 5.0 to 22.5 g/hr.
Within this rectangle there appears to be a random scatter of points, which would
indicate that at this range of burn rates, emission rates can be quite variable.
It is probable that factors other than burn rate (stove design, catalyst
operational characteristics, fueling practices, heat demand, and other factors)
contribute to the resulting particulate emission rates.
Figure 4-2A also presents particulate emission (g/kg) data from the catalytic stove
classification. The r2 values for individual stove models range from 0.069 (Stove
B) to 0.444 (Stove A). Stove P had the lowest r2 value (0.001). The overall r2
value is 0.048, which indicates an extremely poor burn rate/emission rate
correlation.
The majority of data points in Figure 4-2A are grouped in a rectangle bounded by
burn rates of approximately 0.60 to 1.50 kg/hr and emission rates of approximately
5.0 to 20.0 g/kg.
Add-on/Retrofits. Figure 4-1B presents the burn rate (kg/hr) and particulate
emission (g/hr) data from the add-on/retrofit technology classification. The r2
values for individual models range from 0.219 (Add-on J) to 0.832 (Add-on G). The
overall r2 value is 0.003, which indicates a very poor burn rate/emission rate
correlation.
The data points plotted in this figure are all contained within a rectangle bounded
by a burn rate range of 0.8 to 2.4 kg/hr and an emission rate range of 5.0 to 38.0
4-11
-------
g/hr. The points from Add-on I are all located in the higher burn rate (greater
than 2.0 kg/hr) section of the rectangle.
According to conventional wisdom, the relationship of burn rate to emission rate in
add-on/retrofits would be expected to result in relatively high emission rates at
lower burn rates due to difficulties in maintaining catalyst lightoff. At mid-
range burn rates the lowest emission rates would be expected. At high burn rates,
emissions would be expected to increase due to decreased flue gas residence time in
the catalyst. If the data points from Add-on I are eliminated, the remaining data
points appear to outline a rough "U" shape, which is the expected pattern based on
conventional theory. It should be noted that the performance of add-on/retrofit
technology is quite dependent on the characteristics of the conventional stove
model on which the technology is mounted.
Figure 4-2B presents the burn rate (kg/hr) and particulate emission (g/kg) data
from the add-on/retrofit classification. The r2 values for individual models range
from 0.213 (Retrofit E) to 0.695 (Add-on G). The overall r2 value is 0.242, which
indicates a poor burn rate/emission rate correlation.
As in the case of Figure 4-1B, if the Add-on I data points are eliminated from the
data set, the remaining data points appear to outline a rough "U" shape, as would
be anticipated based on conventional theory.
Low-emission Stoves. Figure 4-1C presents the burn rate (kg/hr) and particulate
emission (g/hr) data from the low-emission stove classification. The r2 values for
individual models range from 0.002 (Stove N) to 0.811 (Stove K). The overall r2
value is 0.095, which indicates a very poor burn rate/emission rate correlation.
There is one data point, (1.26,47.6), which appears to be an outlier to the data
set; however, this point is from the data set with the highest r2 value (0.811,
Stove K). The remaining data points appear to be roughly grouped into a isosceles
triangle shape. Within this triangle the majority of data points appear to be
located in the area of burn rates less than 1.0 kg/hr. The location of this group
of points represents burn rates that are significantly lower than the observed burn
rates for the other stove technologies in the study. The data in this subsection
of the triangle indicates that for the low-emission stove models operated at burn
rates of 0.7 to 1.0 kg/hr, emission rates can be quite variable within a range of
3.0 to 17.0 g/hr. As in the case of catalytic stoves, it is probable that other
4-12
-------
stove operation factors, in addition to burn rate, determine the particulate
emission rate in this range of burn rates.
The burn rate to emission rate relationship in low-emission stoves is difficult to
predict, primarily due to the influence of significant variables in the design of
individual models of low-emission stoves (air inlets, flue gas flow patterns,
baffling, etc.), and the potential effect of these variables on burn rates and
emission rates. It appears from Figure 4-1C that no apparent trend is indicated.
However, high burn rate capabilities (greater than 1.4 kg/hr) appear to be lower
for low-emission stoves relative to catalytic stoves, add-on/retrofits, and
conventional stoves.
Figure 4-2C presents burn rate (kg/hr) and particulate emission (g/kg) data from
the low-emission stove classification. The r2 values for individual models range
from 0.044 (Stove N) to 0.751 (Stove K). The overall r2 value is 0.012, which
indicates a very poor burn rate/emission rate correlation.
As in the case of Figure 4-1C, there is a single point from Stove K which appears
to be an outlier (1.26,37.9); however, this point is a part of the data set with
the highest r2 value. If this point is eliminated from the data set and the outer
points of the remaining data grouping are connected, a rough triangle is formed.
The data points appear to be scattered throughout this triangle; however, as in the
case of Figure 4-1C, the majority of points are located in the area of burn rates
less than 1.0 kg/hr. The data pattern in this subsection of the triangle indicates
that for burn rates of 0.6 to 1.0 kg/hr, gram-per-kilogram emission rates can be
quite variable. In this subsection of the triangle, stove operational factors
other than burn rate probably influence the emission rate.
Conventional Stoves. Figure 4-1D presents the burn rate (kg/hr) and particulate
emission (g/hr) data from the conventional stove classification. The overall r2
value for this plot is 0.034, which indicates a very poor burn rate/emission rate
correlation.
There are two data points which appear to be outliers, (1.60,4.7) and (2.45,2.9).
The emission rates associated with these points are significantly less than the
overall emission rate for conventional stoves of 20.1 g/hr. Both points are from
the same home and stove.
4-13
-------
The conventional theory relating emission rate to burn rate for conventional stoves
states that participate emission rate should decrease as burn rate increases. This
is based on lab test data and is assumed to be from elevated firebox temperatures
and turbulence at higher burn rates. The data grouping in Figure 4-1D does not
appear to support this general theory. It is difficult to define a trend based on
the shape of the data plot; however, higher emission rates (greater than 30.0 g/hr)
were not observed at the lower burn rates (less than 1.5 kg/hr). Additionally, the
lower emission rates (less than 20.0 g/hr) were not observed at the higher burn
rates (greater than 1.9 kg/hr).
Figure 4-2D presents the burn rate (kg/hr) and particulate emission (g/kg) data
from the conventional stove classification. The overall r2 value is 0.164, which
indicates a poor burn rate/emission rate correlation.
There are two points (again, from the same home and stove) with emission rates that
are significantly lower than the overall conventional stove gram-per-kilogram
emission rate of 12.1 kg/hr ((1.60,2.9) and (2.45,1.2). If these points are
eliminated the remaining points illustrate a fairly horizontal pattern (relatively
narrow emission rate range accompanied by a relatively wide burn rate range).
These points are grouped in a rectangle bounded by burn rates of approximately 0.90
to 2.5 kg/hr and emission rates of approximately 8.0 to 23.0 g/kg.
Discussion by Stove Model
The following discussion is primarily limited to observations regarding the effect
of burn rate on particulate emission rates as measured during the study. It is
recognized that burn rate is only one of many variables that affect the emission
rate; however, for some stove designs a general relationship of burn rate to
emission rate is discernible. When making performance comparisons between stove
technologies or individual stove models, all factors which influence emission rates
should be considered. The factors of heat demand, fueling practices by the stove
operator, chimney type, stove technology, and any other significant factors should
be considered when comparing a particular stove model with other stove models or
technologies.
The data collection and calculation methods used in this study have estimated
precision and accuracy values attached to each measurement (refer to Appendix C).
Because of the uncertainty associated with each measurement, caution should be used
when comparing technologies or individual stove models. Relatively small
4-14
-------
differences in measured rates or percentages indicate that the numbers may be
virtually identical given the uncertainties associated with each measurement.
In the following discussion the major emphasis is placed on gram-per-hour emission
rates rather than gram-per-kilogram emission rates, primarily because the industry
standard for certification of woodstoves is based on gram-per-hour emission rates.
The relationship of burn rate to emission rates (kg/hr x g/kg = g/hr) should be
remembered when considering gram-per-hour data.
Catalytic Stoves
Stove A. Catalytic Stove A exhibited a mean particulate emission rate range by
home of 18.0 to 22.8 g/hr (two homes), with an overall mean emission rate of 20.4
g/hr. The mean burn rates by home for Stove A ranged from 0.70 to 1.42 kg/hr, with
an overall mean burn rate of 1.10 kg/hr. Stove A had an overall mean emission rate
that was 3.8 g/hr higher than the overall mean for all catalytic stoves of 16.6
g/hr. The Stove A mean burn rate was 0.07 kg/hr lower than the overall mean burn
rate for all catalytic stoves of 1.17 kg/hr.
In Figure 4-1A, the burn rate/emission rate data points for Stove A have an r^
value of 0.118, which indicates a poor emission rate/burn rate correlation. Seven
of the eight data points for Stove A are grouped in a rectangle bounded by burn
rates of approximately 0.55 to 1.70 kg/hr, and emission rates of approximately 10.0
to 24.0 g/hr. The remaining data point, (1.58,39.7), lies outside of this
rectangle.
The data points for stove A appear to be concentrated at either end of the range of
burn rates; no points are located in the burn rate range of 0.8 to 1.4 kg/hr. This
separation is an artifact of the stove operating practices in the two homes where
Stove A was evaluated. The range of burn rates from individual sampling periods in
Home N01 was 0.57 to 0.78 kg/hr, while the range of emission rates in this home was
13.0 to 21.2 g/hr. The range of burn rates from individual sampling periods in
Home N10 was 1.46 to 1.69 kg/hr, while the range of emission rates was 9.7 to 39.7
g/hr.
Although the two homes that used Stove A operated the stove significantly
differently (as indicated by the difference in burn rates), there is no
relationship evident between burn rate and emission rate. Fuel moisture was
similar in both homes. Apparently other factors in addition to burn rate influence
the emission rate in the range of burn rates observed in this data set.
4-15
-------
Stove B. Catalytic Stove B exhibited a mean participate emission rate range by
home of 6.5 to 30.7 g/hr (four homes), with an overall mean emission rate of 20.5
g/hr. The mean burn rates by home for catalytic Stove B ranged from 0.88 to 1.38
kg/hr, with an overall mean burn rate of 1.21 kg/hr. Catalytic Stove B had an
overall mean emission rate that was the highest overall mean rate of all catalytic
stove models, and 4.1 g/hr higher than the overall mean for all catalytic stoves of
16.4 g/hr. The Stove B mean burn rate was 0.04 kg/hr lower than the overall mean
burn rate for all catalytic stoves of 1.17 kg/hr.
In Figure 4-1 A, the burn rate/emission rate data points for Stove B have an r2
value of 0.348, which indicates a poor emission rate/burn rate correlation. Five
of the seven data points plotted in Figure 4-1A with emission rates higher than
27.0 g/hr are points from Stove B, which indicates that this stove design is
capable of relatively high emission rates. Of the five data points mentioned
above, three of the four homes where Stove B was evaluated are represented.
In contrast to the three homes with samples above 27.0 g/hr, one home (Vll) had a
narrow range of relatively low emission rates with Stove B (mean of 6.5 g/hr, range
of 6.1 to 7.0 g/hr, three values). The range of burn rates in Home Vll (1.02 to
1.19 kg/hr) are in the middle of the range for Stove B, so burn rate does not
appear to be a significant influence on the relatively low emission rates observed
in this home. Fuel moistures from Vll were similar to or higher than the other
Stove B homes.
As indicated by the data from Home Vll, Stove B is capable of producing relatively
low emission rates (less than 7.0 g/hr). However, as indicated by the remainder
of the data set on this stove, the stove can be capable of producing relatively
high emission rates. No relationship of burn rate to emission rate was
discernible.
Stove C. Catalytic Stove C exhibited a mean particulate emission rate range by
home of 9.4 to 17.1 g/hr (three homes), with an overall mean emission rate of 14.2
g/hr. The mean burn rates by home for catalytic Stove C ranged from 1.00 to 1.60
kg/hr, with an overall mean burn rate of 1.31 kg/hr. Stove C had an overall mean
emission rate that was 2.4 g/hr lower than the overall mean for all catalytic
stoves of 16.6 g/hr. The Stove C mean burn rate was 0.14 kg/hr higher than the
overall mean burn rate for all catalytic stoves of 1.17 kg/hr.
4-16
-------
In Figure 4-1A, the burn rate/emission rate data points for Stove C have an r^
value of 0.229, which indicates a poor emission rate/burn rate correlation. The
data set from Stove C contains one point, (1.45,1.7), which appears to be an
outlier because of the low emission rate. When this point is eliminated from the
data set, the remaining data points for Stove C still do not appear to show any
discernible relation between burn rate and emission rate. The majority of data
points for Stove C are grouped in a rectangle bounded by burn rates of 0.9 to 1.5
kg/hr and emission rates of 7.0 to 25.0 g/hr.
Home V07 had the lowest mean emission rate (9.4 g/hr) accompanied by the highest
mean burn rate (1.60 kg/hr). Home V16 had the middle mean emission rate (16.2
g/hr) and middle mean burn rate (1.21 kg/hr). Home N03 had the highest mean
emission rate (17.1 g/hr) and the lowest mean burn rate (1.00 kg/hr). The above
relationship consists of a relatively small data set (three homes) and uses mean
values for the data, so the apparent relationship of decreased emissions
accompanied by increased burn rate for Stove C could be a statistical artifact.
This apparent trend contradicts the conventional assumption of the relationship of
burn rate to emission rate in catalytic stoves. Therefore, this observed trend for
Stove C may be a result of influence of factors other than burn rate on emission
rate. Fuel moisture does not correlate with emissions from the three homes.
Stove D. Catalytic Stove D exhibited a mean participate emission rate range by
home of 8.4 to 14.4 g/hr (four homes), with an overall mean emission rate of 12.2
g/hr. The mean burn rates by home for catalytic Stove D ranged from 0.89 to 1.11
kg/hr, with an overall mean burn rate of 1.02 kg/hr. Catalytic Stove D had an
overall mean emission rate that was the lowest of all catalytic stoves and 2.0 g/hr
lower than the overall mean for all catalytic stoves of 16.4 g/hr. The Stove D
mean burn rate was also the lowest of all catalytic stoves and 0.15 kg/hr lower
than the overall mean burn rate for all catalytic stoves of 1.17 kg/hr.
In Figure 4-1A, the burn rate/emission rate data points for Stove D have an r^
value of 0.051, which indicates a very poor emission rate/burn rate correlation.
The data points for Stove D appear to be randomly located in a rectangle bounded by
a burn rate range of 0.6 to 1.3 kg/hr and an emission rate range of 5.0 to 21.0
g/hr. This grouping of data points indicates that Stove D is capable of relatively
consistent emissions performance (under a relatively narrow range of burn rates)
compared to the other catalytic stoves evaluated in the study.
4-17
-------
The most significant factor in the performance of Stove D appears to be the
replacement of cordierite-based combustors with mullite-based combustors.
Emissions from all Stove D units decreased during the second heating season after
the mullite-based combustors were installed. It is unclear whether this
improvement is due to better combustion performance by the new combustors or simply
because the combustors were not degrading as earlier units did. Another
significant factor that separates Stove D from the other catalytic stoves is
firebox size. The Stove D firebox volume is 38 liters, which is smaller than the
other catalytic stove firebox volumes (87 liters for Stove A, 122 liters for Stove
B, and 69 liters for Stove C). The smaller firebox size may have tended to limit
burn rates in Stove D relative to the other catalytic models, a situation which is
similar to that observed for the low-emission stoves, which had relatively low burn
rates and relatively small fireboxes. The smaller firebox in Stove D may have
contributed to the lower overall mean emission rate due to lower emissions during
periods when the catalyst was not operational.
Stove D did not appear to exhibit a discernible relationship between burn rate and
emission rate; however, the stove design (smaller firebox) may tend to limit burn
rate and emission rates. Fuel moisture was low (about 15% DB) in N02 and Nil.
Stove Code P. The category of "Stove Code P" included four homes which had
purchased catalytic stoves prior to the commencement of the project. These "pre-
existing" catalytic stoves (different models for each home) were evaluated to give
an indication of catalytic stove performance after three or more seasons of use.
The data set for Stove Code P is relatively small (one sampling period for Homes
V31 and N32, two sampling periods for Homes V32 and N33).
In Figure 4-1A, the burn rate/emission rate data for Stove Code P have an r2 value
of 0.811, which is the highest value for all catalytic stove models. This r2 value
should be considered an artifact of a relatively small data set (six values),
because four catalytic stove models are represented under Stove Code P- Four of
the six data points for Stove Code P are roughly grouped with other catalytic stove
data. Two data points ([1.83,22.3] and [2.26,34.6], both recorded in Home N33)
exhibit relatively high burn rates, and one of them, (2.26,34.6), exhibits a
relatively high emission rate.
Add-on/Retrofits
Retrofit E. The data set for Retrofit E is relatively small, as only one home
(V01) was evaluated, and emission data from three sampling periods were obtained.
4-18
-------
Retrofit E exhibited a participate emission rate range for individual sampling
periods in Home V01 of 6.3 to 10.1 g/hr, with an overall mean emission rate of 7.8
g/hr. The mean burn rates ranged from 1.17 to 1.58 kg/hr, with an overall mean
burn rate of 1.37 kg/hr. The Retrofit E mean emission rate (based on one home) was
the lowest mean emission rate observed for all stove models evaluated in the study,
and 10.4 g/hr lower than the overall mean for all catalytic retrofits of 18.2 g/hr.
The Retrofit E mean burn rate was 0.12 kg/hr higher than the overall mean burn rate
for all catalytic retrofits of 1.25 kg/hr.
In Figure 4-1B, the r^ value for Retrofit E was 0.396, which indicates a poor to
fair emission rate/burn rate correlation for the three data points (although three
data points do indicate a general trend of increased emission rate accompanied by
increased burn rate).
In this retrofit design, the combustor assembly is placed inside the firebox,
resulting in a combustor arrangement similar to that of other integrated-catalyst
stoves. The three data points for Retrofit E are grouped relatively closely which
is an indication that the retrofit is capable of achieving relatively low-emission
rates under the range of burn rates observed in this study. Conclusions on the
effectiveness of Retrofit E at reducing emission rates should be made with caution
due to the relatively small data set from only one home. The data does indicate
that as operated in Home V01, Retrofit E is capable of achieving low particulate
emission rates.
Retrofit F. Retrofit F exhibited a mean particulate emission rate range by home of
25.2 to 26.6 g/hr (two homes), with an overall mean emission rate of 25.9 g/hr.
The mean burn rate range by home for Retrofit F was 1.10 to 1.23 kg/hr, with an
overall mean burn rate of 1.15 kg/hr. The overall mean particulate emission rate
for Retrofit F is the highest overall mean emission rate observed for all add-on/
retrofit devices and 7.7 g/hr higher than the overall mean for all catalytic
retrofits of 18.2 g/hr. The Retrofit F mean burn rate was 0.10 kg/hr lower than
the overall mean burn rate for all catalytic retrofits of 1.25 kg/hr.
In Figure 4-1B, the r^ value for Retrofit F was 0.273, which indicates a poor
emission rate/burn rate correlation. For three of the four data points, a
relatively narrow range of burn rates was measured (0.87 to 0.97 kg/hr),
accompanied by a relatively wide range of emission rates (16.5 to 36.7 g/hr). The
two relatively high emission rates were measured in each of the two homes that used
Retrofit F (31.8 g/hr in Home V03, 36.7 g/hr in Home V12). This is an indication
4-19
-------
that the emission rate performance of the retrofit is probably affected by other
factors in addition to burn rate. Fuel moisture was relatively low in both homes.
Add-on G. Add-on G exhibited a mean particulate emission rate range by home of
16.3 to 18.7 g/hr (two homes), with an overall mean emission rate of 17.1 g/hr.
The mean burn rate range by home for Add-on G was 1.62 to 1.78 kg/hr, with an
overall mean burn rate of 1.70 kg/hr. The Add-on G mean emission rate was 0.6 g/hr
lower than the overall mean for all catalytic add-ons of 17.7 g/hr- The Add-on G
mean burn rate was 0.05 kg/hr lower than the overall mean burn rate for all
catalytic add-ons of 1.75 kg/hr.
In Figure 4-1B, the r2 value for Add-on G was 0.832, which was the highest r2 value
for all add-on/retrofits. The data set for Add-on G is fairly small (three
values); however, the data does indicate a general trend of increased emissions
with an increased burn rate. The burn rate range (1.61 to 1.70 kg/hr) and emission
rate range (15.5 to 18.7 g/hr) for Add-on G are both relatively narrow, which could
account for the apparent burn rate/emission rate correlation.
Caution should be used in attempting to establish an emission rate/burn rate
correlation for Add-on G. It appears that the relatively narrow ranges of emission
rate values and burn rate values in addition to the small data set make a
determination of an emission rate/burn rate correlation difficult.
Add-on H. Add-on H was evaluated for one sampling period in two homes (V10 and
V15); however, only one particulate emission sample was obtained due to an
equipment malfunction in Home V15. This single sample indicated an emission rate
of 16.2 g/hr and a burn rate of 1.01 kg/hr.
Comparisons of Add-on H with other devices in the add-on/retrofit category are
difficult to make due to the single sample. As there is only one data point, there
is no burn rate/emission rate trend evident.
Add-on I. Add-on I exhibited a mean particulate emission rate range by home of
22.6 to 25.7 g/hr (two homes), with an overall mean emission rate of 23.4 g/hr.
The mean burn rate range by home for Add-on I was 2.23 to 2.35 kg/hr, with an
overall mean burn rate of 2.26 kg/hr. The Add-on I mean emission rate was the
highest observed for all add-ons, and 5.7 g/hr higher than the overall mean for all
catalytic add-ons of 17.7 g/hr. The Add-on I mean burn rate was the highest mean
4-20
-------
burn rate observed for all add-ons and 0.48 kg/hr higher than the overall mean burn
rate for all catalytic add-ons of 1.75 kg/hr.
In Figure 4-1B, the r2 value for Add-on I was 0.586, which indicates a fair
emission rate/burn rate correlation. The relatively high burn rates observed for
Add-on I were recorded in two homes which used different conventional stove models
but operated at similar burn rates. Because the range of burn rates for measured
for individual sampling periods with Add-on I was relatively narrow (2.08 to 2.35
kg/hr), the plot of burn rate vs. emissions in Figure 4-1B shows a relatively wide
range of emission rates (13.6 to 37.3 g/hr) associated with a relatively narrow
range of burn rates. This would appear to indicate that other factors in addition
to burn rate affects the emission rates in the range of burn rates observed. Fuel
moisture was low in one home (about 22%) and high (about 42%) in another.
Add-on J. Add-on J exhibited a mean particulate emission rate range by home of 7.3
to 14.9 g/hr (three homes), with an overall mean emission rate of 12.8 g/hr. The
mean burn rate range by home for Add-on J was 1.27 to 1.78 kg/hr, with an overall
mean burn rate of 1.46 kg/hr. The Add-on J mean emission rate was the lowest mean
emission rate observed for all add-ons, and 4.9 g/hr lower than the overall mean
for all catalytic add-ons of 17.7 g/hr. The Add-on J mean burn rate was 0.29 kg/hr
lower than the overall mean burn rate of all catalytic add-ons of 1.75 kg/hr-
The data from Add-on J in Figure 4-2B has an r2 value of 0.219, which indicates a
poor emission rate/burn rate correlation. Fuel moisture from Add-on J samples was
relatively low (20-25%).
Low-emission Stoves
Low-emission Stove K. Low-emission Stove K exhibited a mean particulate emission
rate range by home of 11.2 to 29.5 g/hr (two homes), with an overall mean emission
rate of 23.4 g/hr. The mean burn rate range by home for Stove K was 0.86 to 1.13
kg/hr, with an overall mean burn rate of 1.02 kg/hr. The Stove K mean emission
rate was the highest mean emission rate observed for all low-emission stoves and
10.0 g/hr higher than the overall mean for all low-emission stoves of 13.4 g/hr.
The Stove K mean burn rate was 0.02 kg/hr higher than the overall mean burn rate of
all low-emission stoves of 1.00 kg/hr.
The data points for Stove K in Figure 4-2C have an r2 value of 0.811, which is the
highest r2 value for all low-emission stoves. The data points for Stove K exhibit
a fairly linear trend which shows a relatively sharp increase in emission rate
4-21
-------
accompanied by a relatively small change in burn rate. For individual sampling
periods, Stove K had an emission rate range of 9.4 to 47.6 g/hr, and a burn rate
range of 0.84 to 1.26 kg/hr.
The emissions performance of Stove K was significantly different in the two homes
where it was evaluated (V18 and N07). In Home V18, the emission rates for four
individual sampling periods ranged from 17.3 to 47.6 g/hr. In Home N07, the
emission rates for two individual sampling periods were 9.4 and 12.9 g/hr.
A significant operational difference between the two homes is reflected in the
observed burn rates. In Home V18 the burn rates for individual sampling periods
ranged from 1.08 to 1.26 kg/hr. In Home N07 the burn rates for individual sampling
periods ranged from 0.84 to 0.90 kg/hr.
The influence of burn rate on particulate emission rate magnitude for Stove K is
dramatically demonstrated by the data set collected in Home V18. During the first
three sampling periods in this home, emission rates ranged from 17.3 to 28.3 g/hr,
while burn rates ranged from 1.08 to 1.10 kg/hr. During the fourth sampling period
in this home, the highest burn rate (1.26 kg/hr) was measured accompanied by the
highest emission rate (47.6 g/hr). Fuel moisture averaged more than 30% in Home
V18 and around 20% in Home N07.
Low-emission Stove L. Low-emission Stove L exhibited a mean particulate emission
rate range by home of 9.2 to 9.6 g/hr (two homes), with an overall mean emission
rate of 9.4 g/hr. The mean burn rate range by home for Stove L was 0.90 to 1.15
kg/hr, with an overall mean burn rate of 1.01 kg/hr. The Stove L mean emission
rate was 4.0 g/hr lower than the overall mean for all low-emission stoves of 13.4
g/hr. The Stove L mean burn rate was 0.01 kg/hr higher than the overall mean burn
rate of all low-emission stoves of 1.00 kg/hr.
The data points in Figure 4-2C have an r2 value of 0.120, which indicates a poor
emission rate/burn rate correlation. The data points appear to exhibit a "flat"
pattern (relatively narrow range of emission rates accompanied by a relatively
large range of burn rates). It appears that Stove L is capable of sustaining
relatively low emission rates (range of 6.5 to 14.1 g/hr for individual sampling
periods) under a relatively wide range of burn rates (range of 0.76 to 1.34 kg/hr
for individual sampling periods). Fuel moistures were very low in both homes,
averaging about 14% for V04 and 16% for N15.
4-22
-------
Low-emission Stove M. Low-emission Stove M exhibited a mean participate emission
rate range by home of 6.9 to 21.8 g/hr (three homes), with an overall mean emission
rate of 12.5 g/hr. The mean burn rate range by home for Stove M was 0.67 to 1.18
kg/hr, with an overall mean burn rate of 0.91 kg/hr. The Stove M mean emission
rate was 0.9 g/hr lower than the overall mean for all low-emission stoves of 13.4
g/hr. The Stove M mean burn rate was 0.09 kg/hr lower than the overall mean burn
rate of all low-emission stoves of 1.00 kg/hr.
The data points in Figure 4-2C have an r2 value of 0.597, which indicates a fair
emission rate/burn rate correlation. Four of the five data points appear to
exhibit a general trend of increased emission rate accompanied by increased burn
rate.
The two highest emission rates for Stove M (17.2 and 26.3 g/hr) were measured in
Home V14. It is apparent that emissions in Home V14 were significantly higher than
in Homes V12 (5.2 g/hr, one sampling period) and V34 (5.9 and 7.9 g/hr, two
sampling periods). Fuel moisture was not significantly different between the
homes, averaging between about 20% and 30%.
Stove M was added to the study before the start of the 1986-87 heating season, and
was considered "state-of-the-art" low-emission stove technology. Laboratory
testing on this stove has indicated that it is capable of meeting EPA 1990 NSPS
emission standards.
Low-emission Stove N. Low-emission Stove N exhibited a mean particulate emission
rate range by home of 3.6 to 10.2 g/hr (three homes), with an overall mean emission
rate of 8.1 g/hr. The mean burn rate range by home for Stove N was 0.90 to 1.33
kg/hr, with an overall mean burn rate of 1.08 kg/hr. The Stove N mean emission
rate was the lowest of all low-emission stoves and 5.3 g/hr lower than the overall
mean for all low-emission stoves of 13.4 g/hr. The Stove N mean burn rate was 0.08
kg/hr higher than the overall mean burn rate of all low-emission stoves of 1.00
kg/hr.
The data points in Figure 4-2C have an r2 value of 0.002, which indicates a very
poor emission rate/burn rate correlation. The six data points plotted are
scattered within a rectangle bounded by burn rates of approximately 0.75 to 1.40
kg/hr and emission rates of approximately 2.0 to 19.0 g/hr.
4-23
-------
The three lowest emission rates measured were recorded in one of each of the three
homes using Stove N (2.0 g/hr in Home V03, 3.6 g/hr in Home V35, and 4.3 g/hr in
Home N16). This is an indication of the ability of Stove N to achieve relatively
low emission rates in the three different in-situ installations in this study. The
poor correlation of burn rate with emission rate for Stove N indicates that factors
other than burn rate influence the emission rate for this stove design. Fuel
moisture ranged between 25% and 35% in the three homes.
Stove N was also added to the study before the start of the 1986-87 heating season,
and was considered "state-of-the-art" low-emission stove technology. It also has
been shown to be capable of meeting EPA 1990 standards.
FUELING EFFECTS
Fuel Loading Frequency Effects on Particulate Emissions
The following discussion is limited to apparent effects of fuel loading frequency
on particulate emission rate. It is recognized that several factors in addition to
fuel loading frequency affect the particulate emission rate; however, the purpose
of this discussion is to determine whether or not a fuel loading frequency/emission
rate relationship is evident for individual stove models or stove technology
classifications.
Conventional wisdom suggests that relatively short, hot fires (an operation style
which would be expected to be reflected by increased fuel loading frequency)
decrease particulate emissions.
Figures 4-3A through 4-3D are X-Y plots of fuel loading frequency (#/hr) versus
particulate emission rate (g/hr) from individual sampling periods for each of the
four stove technologies evaluated (catalytic stoves, add-on/retrofits, low-
emission stoves, and conventional stoves). For each data set, linear regression
coefficients (r2 values) were calculated for each stove model with three or more
valid data points and for the total data set in each technology category.
Catalytic Stoves. Because the design of all of the catalytic stove models in this
study incorporates a bypass damper, a hypothesis might be made that emission rates
would increase with fueling frequency due to a higher percentage of time that the
bypass damper is in use (catalyst disengaged) and increased potential for losing
catalyst "light-off" temperatures. However, the data from this study does not
indicate a discernible fueling frequency/emission rate correlation.
4-24
-------
4=>
I\D
cn
Figure 4-3A
Particulate Emissions (g/hr) vs. Fuel Loading Frequency (ft/hr)
Catalytic Stoves
fS -
fO -
35 -
PflRTICULflTE
EMISSIONS
(G/HR) '
25 -
20 -
15 -
10 -
5 -
o
D X
• ' ,
D
o
x i.
• 0 •
D
°'°5
°'10
0.20 0.25 0.30
FUEL LDflDING FREQUENCV (tf/HR)
0,35
O.tO
O.H5
0.50
ALL (R2 =0.001)
0 STDUE ft (R2 =0.135)
O STDUE B (R2 =0.159)
D STDUE C U2 =0.102)
| STDUE D (R2 =0.002)
X SIDUE F (R2 =0.205)
-------
Figure 4-3B
Participate Emissions (g/hr) vs. Fuel Loading Frequency (ft/hr)
Add-On/Retrofits
ro
01
to -
35 -
PflRTICULflTE
EHISSIDNS
(G/HR) '
25 -
20 -
15 -
10 -
5 -
D
5
o
YD X
X
0.05
0.10
0,15
flLL (R2 =0,023)
0 STDUE E (R2 =0.t39)
O STDUE G (R2 =0 .923)
0.20 0.25 0.30 0.35
FUEL LOfiDING FREQUEHCV (ff/HR)
D SIDUE F (R2 =0.313)
I STDUE J (R2 =0.000)
X STDUE I CR2 =0.006)
O.HO
0 H5
0.50
Y STDUE H (R2 = Hfl )
-------
Figure 4-3C
Particulate Emissions (g/hr) vs. Fuel Loading Frequency (tt/hr)
Low-Emission Stoves
fO -
35 -
PflRTICULflTE
EMISSIONS
(G/HR5 '
25 -
20 -
15 -
10 -
5 -
D
D
0
D
0.05 0.10 0.15 0.20 0.25 0.30
FUEL LDftDING FREOUENCV (0/HR)
0.35
O.HO
O.fS
0.50
ftLL (R2 =0.002)
0 STDUE N (R2 =0.352)
O STDUE L (R2 =0.0f9)
D SIDUE II (R2 =0.013)
| STDUE K (R2 =0.t55)
-------
Figure 4-3D
Particulate Emissions (g/hr) vs. Fuel Loading Frequency (Jt/hr)
Conventional Stoves
35 -
PflRTICLJLflTE
EHISSIDNS
(G/HR)
r\3
00
25 -
10 -
C r__
i
I
i
1
I
I
0.05
0.10
0,15
0,20 0.2E 0.30 0,JE
FUEL LDflDING FREQUENCY (fl/HR)
O.tO
O.tS
0,50
-------
Figure 4-3A presents the fueling frequency/emission rate data for the catalytic
stove classification. The r2 values for individual stove models range from 0.002
(Stove D) to 0.195 (Stove B). The overall r2 value of 0.001 indicates a very poor
fueling frequency/emission rate correlation.
As previously discussed in the burn rate/emission rate analysis, the emission rates
for the catalytic stoves can be quite variable. Consequently, there is a wide
range of emission rates represented in Figure 4-3A (1.7 to 41.3 g/hr). It appears
that the majority of data points are located in the fueling frequency range of 0.10
to 0.25 #/hr.
There are 11 data points which are located in the area of fueling frequencies
greater than 0.25 #/hr and emission rates less than 23.0 g/hr. In contrast to this
observation, there is only one data point located in this fueling frequency range
with an emission rate greater than 23.0 g/hr. This may appear to indicate a
correlation of lower emission rates with higher fueling frequency; however, it is
more likely that this observation is an artifact of the catalytic stove data set
(there are relatively few points with emission rates greater than 23.0 g/hr).
Two data points are located in the area of fueling frequency less than 0.10 #/hr
(both points from Stove B). This indicates that the fueling cycle for Stove B, a
large stove, can be as long as once every 10 hours under "typical" Northeast
conditions.
None of the individual stove models appear to have a good fueling frequency/
emission rate correlation (r2 = 0.321). The pattern of data points is fairly
nonconclusive.
Add-on/Retrofits. Figure 4-3B presents the fueling frequency/emission rate data
for the add-on/retrofit classification. The r2 values for individual models range
from 0.000 (Add-on J) to 0.949 (Retrofit F). The overall r2 value is 0.023, which
indicates a very poor fueling frequency/emission rate correlation.
The data points from the overall add-on/retrofit classification do not appear to
exhibit any general fueling frequency/emission rate relationship. The points are
all located in a rectangle bounded by emission rates of approximately 5.0 to 38.0
g/hr and by fueling frequencies of approximately 0.11 to 0.38 #/hr.
4-29
-------
Over 50% of the data points appear to be located in a rectangle bounded by emission
rates of 5.0 to 19.0 g/hr and fueling frequencies of 0.18 to 0.27 #/hr. This area
represents a higher fueling frequency than indicated for the catalytic stoves (the
majority of data points for catalytic stoves were bounded by a low fueling
frequency of 0.10 #/hr).
Retrofit F (four data points) and Add-on G (three data points) had very good r2
values (0.949 and 0.923, respectively). Retrofit F demonstrated a trend of
increased particulate emission rate with decreased fueling frequency. Add-on G
demonstrated a trend of increased particulate emission rate with increased fueling
frequency. Caution should be used when interpreting these apparent trends due to
the small sample populations and the lack of an apparent overall trend from the
add-on/retrofit technology classification.
Low-emission Stoves. Figure 4-3C presents the fueling frequency/emission rate data
for the low-emission stove classification. The r2 values for individual models
range from 0.013 (Stove M) to 0.452 (Stove K). The overall r2 value is 0.002,
which indicates a very poor fueling frequency/emission rate correlation.
As indicated in the emission rate/burn rate discussion, the emission rate from the
low-emission stoves can be quite variable. All but four of the data points in
Figure 4-3C are located in the fueling frequency range of 0.23 to 0.36 #/hr.
Although the particulate emission rates vary considerably within this fueling
frequency range, the high percentage of the data set located in this range
indicates that the design of the low-emission stoves appears to dictate a
prescribed fueling frequency range. This could be due to the need to refuel (to
maintain stove operation), but the inability to load large amounts of fuel (due to
smaller firebox sizes).
The range of fueling frequencies for the low-emission stove data set is relatively
narrow, making it difficult to discern any fueling frequency/emission rate trends
within this range.
Stove K had the highest r2 value (0.455), and appears to exhibit a curve indicating
increased emission rate accompanied by increased fueling frequency. Two of the
four data points previously mentioned that fall outside of the fueling frequency
range of 0.23 to 0.36 #/hr are from the Stove K data set. These two points give
definition to the Stove K curve; the four remaining data points for Stove K fall
within the fueling frequency range of 0.23 to 0.36 #/hr. This trend for Stove K
4-30
-------
should be interpreted with caution because the majority of data points for Stove K
fall in the 0.23 to 0.36 #/hr fueling frequency range.
The remaining two points that fall outside of the 0.23 to 0.36 l/hr fueling
frequency range are from Stove L. Stove L has a relatively narrow range of
associated emission rates (6.5 to 14.1 g/hr) associated with a relatively wide
range of fueling frequencies (0.31 to 0.53 #/hr). This is an indication that the
emission rate from Stove L may be fairly insensitive to fueling frequency. This
observation should be interpreted with caution because only two of the six data
points for Stove L fall outside of the fueling frequency range of 0.23 to 0.36
#/hr.
Conventional Stoves. Figure 4-3D presents the fueling frequency/emission rate data
for the conventional stove classification. Conventional stoves were not separated
by stove model. The overall r^ value for the conventional stoves is 0.008, which
indicates a very poor fueling frequency/emission rate correlation.
All but two of the data points from the conventional stoves fall in the fueling
frequency range of 0.23 to 0.37 l/hr. This is approximately the same fueling
frequency range observed for the low-emission stoves. Within this range, the
particulate emission rates for the conventional stoves are quite variable (2.9 to
32.6 g/hr).
Because the majority of the data points for the conventional stoves are located in
a relatively narrow range of fueling frequencies, it is difficult to identify a
fueling frequency/emission rate relationship.
Fuel Loading Frequency Effects on Burn Rate
The following discussion is limited to apparent effects of fuel loading frequency
on burn rate. It is recognized that several factors in addition to fuel loading
frequency affect the burn rate; however, the purpose of this discussion is to
determine whether a fuel loading frequency/burn rate relationship is evident for
individual stove models or stove technology classifications.
Conventional wisdom regarding the relationship of fuel loading frequency to burn
rate suggests that an increased fuel loading frequency should be accompanied by an
increased burn rate. (This assumes that heat output from the stove will be
constant.) Other factors, such as stove design, stove technology, and lifestyle of
the stove operator may tend to complicate this correlation.
4-31
-------
Figures 4-4A through 4-4D are plots of fuel loading frequency (#/hr) versus burn
rate (kg/hr) from individual sampling periods for each of the four stove
technologies evaluated (catalytic stoves, add-on/retrofits, low-emission stoves,
and conventional stoves). For each data set, linear regression coefficients (r2
values) were calculated for each stove model with three or more valid data points
and for the total data set in each technology category.
Catalytic Stoves. Figure 4-4A presents the fueling frequency/burn rate data for
catalytic stoves. The r2 values for individual stove models range from 0.000
(Stove B) to 0.297 (Stove A). The overall r2 value is 0.014, which indicates a
very poor fueling frequency/burn rate correlation.
As in the fueling frequency/emission rate analysis, the majority of data points for
the catalytic stoves are located in the fueling frequency range bounded by
approximately 0.10 to 0.25 #/hr. Within this range, the data points are also
bounded by burn rates of approximately 0.5 to 1.8 kg/hr.
Identification of any fueling frequency/emission rate relationship is difficult.
The majority of data points appear to be randomly located in the previously
described range. The data points that lie outside this rectangle also appear to be
randomly located, and do not indicate any discernible fueling frequency/burn rate
relationship.
The data sets for the individual stove models all have r2 values of less than
0.297, and do not appear to exhibit any fueling frequency/burn rate correlations.
It appears that burn rate is not significantly affected by fueling frequency for
the catalytic technology stoves.
Add-on/Retrofits. Figure 4-4B presents the fueling frequency/burn rate data for
the add-on/retrofit classification. The r2 values for individual models range from
0.007 (Add-on J) to 0.815 (Add-on G). The overall r2 value is 0.331, which
indicates a fair fueling frequency/burn rate correlation.
If the outer data points on Figure 4-4B are connected, a rough ellipse is formed
with the long axis tilted to indicate increased fueling frequency with increased
burn rate.
This general trend is supported by four of the six add-on/retrofit devices.
However, the data set from Add-on J (r2 = 0.007) consists of five data points that
4-32
-------
2.5 -
BURN
RflTE
(KG/-HR)
I
CO
GO
1.5 -
1.0 -
0.5 -
Figure 4-4A
Burn Rate (kg/hr) vs. Fuel Loading Frequency (ft/hr)
Catalytic Stoves
n
D
0 o
D
D
A
0
• I
••
o
I I I I I I
0.05 0.10 0.15 0.20 0.25 0.30
FUEL LDftDING FREQUENCY C0/HR)
0.35
O.tO
0.15
0.50
flLL (R2 =0,014)
^) STDUE fi (R2 =0.297)
O STDUE E (R2 =0.000)
D STDUE C
-------
2.5 -
BURN
RftTE
(KG/HR)
I
CO
1.5-
1.0 -
0.5 -
Figure 4-4B
Burn Rate (kg/hr) vs. Fuel Loading Frequency (ft/hr)
Add-On/Retrofits
D
D
B
0
Y
0
D
D
I
0 .05
0.10
0.15
flLL (R =0
0 STDUE E (R2 =0.505)
o STDUE G (R2 =0 .815)
0.20 0.25 0.30 0.35
FUEL LDflDING FREQUENCV (0/HR3
D STDUE F (R2 =0.22*)
I STDUE J tR2 =0.007)
X STDUE I (R2 =0.117)
O.tO
O.H5
0.50
Y STDUE H (R2 = Nfl
-------
2.5 -
BURN
RflTE
(KG/HR)
i
. oo
en
1.5 -
1.0 -
0.5 -
Figure 4-4C
Burn Rate (kg/hr) vs. Fuel Loading Frequency (ft/hr)
Low-Emission Stoves
B g
o
i
D
I I I I I I
0.05 0.10 0.15 0.20 0.25 0.30
FUEL LDflDING FREQUENCV (fl/HR)
0.35
0.40
0 45
0.50
ALL (R2 =0.215)
0 STDUE N (R2 =0.038)
O STQUE L (R2 =0.836)
D STDUE H (R2 =O.OS4)
| STDUE K (R2 =0.807)
-------
Figure 4-4D
Burn Bate (kg/hr) vs. Fuel Loading Frequency (ft/hr)
Conventional Stoves
2.5
BURN
RATE
(KG/HR)
00
CTl
1,5 -
1.0 -
0,5 -
i
V •
\ \ r I T i i
O.OE 0.10 0.15 0.20 0.2E 0.20 0,55
FUEL LDflDINS FREQUENCY Iff/HK)
O.fO
O.tS
O.EO
-------
are arranged in an irregular pentagon shape, which does not indicate any apparent
fueling frequency/burn rate relationship. The data set from Add-on H consists of
one point, so no fueling frequency/burn rate relationship is evident. The
remaining four add-on/retrofits (Retrofits E and F, Add-ons G, and I) have r2
values in the range of 0.228 to 0.815. Each of the data sets follows the general
trend of increased fueling frequency with increased burn rate.
This trend should be interpreted with caution due to the relatively small number of
data points which are used to determine the trend. For example, if data points
with burn rates higher than 2.0 kg/hr (four points from Add-on I) and data points
with fueling frequencies less than 0.18 #/hr (two points from Retrofit F, two
points from Add-on G) are eliminated, the trend of increased fueling frequency with
increased burn rate is no longer evident.
Low-emission Stoves. Figure 4-4C presents the fueling frequency/burn rate data for
the low-emission stove classification. The r2 values for individual models range
from 0.038 (Stove N) to 0.836 (Stove L). The overall r2 value is 0.215, which
indicates a poor fueling frequency/burn rate correlation.
As previously discussed in the emission rate/burn rate analysis section, the low-
emission stoves exhibited a relatively narrow range of burn rates in comparison
with other stove technologies. In Figure 4-4C, all of the data points are located
within a burn rate range of approximately 0.6 to 1.4 kg/hr. Within this range of
burn rates a relatively wide range of fueling frequencies (0.15 to 0.53 l/hr) was
observed.
It is difficuH to identify an overall fueling frequency/burn rate trend in Figure
4-4C. However, Stoves L (r2 = 0.836, seven points) and K (r2 = 0.807, seven
points) exhibit general trends of increased fueling frequency with increased burn
rate.
Conventional Stoves. Figure 4-4D presents the fueling frequency/burn rate data for
the conventional stove classification. The conventional stoves were not separated
by stove model. The overall r2 value is 0.032, which indicates a poor fueling
frequency/burn rate correlation.
The data points in Figure 4-4D do not exhibit any identifiable fueling
frequency/burn rate relationship. A relatively wide range of burn rates (0.92 to
2.45 kg/hr) is represented; however, all but two of the data points are in a
4-37
-------
fueling frequency range of 0.23 to 0.37 #/hr. Within this fueling frequency range
there is no discernible fueling frequency/burn rate relationship.
Fuel Loading Frequency Effects on Average Fuel Load
The following discussion is limited to apparent effects of fuel loading frequency
on fuel load size. Conventional wisdom regarding the relationship of fueling
frequency to fuel load indicates that as fueling frequency increases, fuel load
size should decrease (assuming a constant burn rate).
Figures 4-5A through 4-5D are plots of fuel loading frequency (#/hr) versus average
fuel load (kg) from individual sampling periods for each of the four stove
technologies evaluated (catalytic stoves, add-on/retrofits, low-emission stoves,
and conventional stoves). For each data set, linear regression coefficients (r2
values) were calculated for each stove model with three or more valid data points
and for the total data set in each technology category.
Catalytic Stoves. Figure 4-5A presents the fueling frequency/fuel load data for
the catalytic stove classification. The r2 values for individual models range from
0.007 (Stove A) to 0.838 (Stove B). The overall r2 value is 0.448, which indicates
a fair fueling frequency/burn rate correlation.
The data points in Figure 4-5A appear to show a trend of decreased fuel loading
frequency accompanied by increased average fuel load. The majority of data points
appear to lie on a curve which starts at approximately (2.0,0.40) and continues
through approximately (14.0,0.05).
The data sets from individual stove models all appear to follow the general trend
of decreased fueling frequency with increased fuel load.
The data set from Stove D has a relatively narrow range of fuel loads (2.4 to 5.8
kg). This is probably a result of Stove D having the smallest firebox of all
catalytic stoves evaluated. The relatively small firebox for this stove would tend
to limit maximum fuel load size.
The Stove P data set has two data points, (6.8,0.27) and (6.9,0.33), that appear to
be outliers from the general trend outlined by the other catalytic stove data
points. The two outlying points represent the data from Home N33.
4-38
-------
LO
VO
Figure 4-5A
Fuel Loading Frequency (8/hr) vs. Average Fuel Load (kg)
Catalytic Stoves
0.50 1
o.ts -
O.fO -
FUEL
LORDING
FREQUENCY
(ff/HR)
0,25 -
0.20 -
O.iS -
0,10 -
0.05 -
B "
X
X
1 •!
j*
XI v X
• 0
• 'jwt oo oD
1 ^o So ^ D08 n 0
° D ° %
0
I 1 1 1 1 1 1
2 t 6 8 10 12 1*
ftUERftGE FUEL LDflO (KG)
ALL (R =
0 SIDUE fl (R2 =0.007)
O STOUE B (R2 =0.838)
D STDUE C (R2 =0,371)
| STDUE D tR2 =0.f29)
X STDUE P (R2 =0.338)
-------
-pa,
I
Figure 4-5B
Fuel Loading Frequency (tt/hr) vs. Average Fuel Load (kg)
Add-On/Retrofits
0.50 -
O.tE -
o.to -
FUEL
LORDING
FREQUENCY
<#/HR)
0,25 -
0.20 -
O.iE -
0,10 -
0.05 -
-1-
X
X
1
1 G
D 0 X X
Y A
0 0 DIo
D
o
D
1 1 1 1 1 1 1
2 4 £ 8 10 12 it
flUERfiGE FUEL LDflD (KG)
ftLL (R2 =0
0 STOUE E (R2 =0 .002)
O STDUE G (R2 =0 .358)
D SIDUE F (R2 =0.00?)
B STDUE J CR2 =0.616)
X STQUE I (R2 =0.837)
Y STDUE H (R2 = Hfl )
+ STDUE D (R2 = Hfl )
-------
Figure 4-5C
Fuel Loading Frequency (ft/hr) vs. Average Fuel Load (kg)
Low-Emission Stoves
0.50 -
O.tS -
(MO -
FUEL
LDflDING
FREQUENCY
(#/HR)
0,2£ -
0.20 -
0, IE -
0,10 -
0.05 -
O
o
0
AJi
QL 0
8 • J A
OLH
no
DD 0
I
I
I
1 1 1 1 1 1 1
2 f 6 8 10 12 If
flUERftGE FUEL LDfiD (KG3
ULl
-------
Figure 4-5D
Fuel Loading Frequency (ft/hr) vs. Average Fuel Load (kg)
Conventional Stoves
0.50 -
0 f5 -
O.fO -
FUEL
LDfiDIMG
FREQUENCY
(tf/HR)
0.25 -
0,20 -
0.15 -
0. 10 -
O.OE -
1
I
I
8 10
FUEL LOAD (KG)
-------
Although Stove A had the lowest r2 value (0.007), the data points from this stove
model do appear to lie within the general trend defined by all catalytic stove data
points. There are seven data points for Stove A that fall within a relatively
narrow fuel load range (4.3 to 5.3 kg) that have a relatively wide fueling
frequency range (0.13 to 0.28 #/hr).
Add-on/Retrofits. Figure 4-5B presents the fueling frequency/fuel load data for
the add-on/retrofit classification. The r2 values for individual models range from
0.002 (Retrofit E) to 0.968 (Add-on G). The overall r2 value is 0.112, which
indicates a poor fueling frequency/burn rate correlation.
The fueling frequency/fuel load relationship is not as clearly delineated as in the
case of the catalytic stoves. The data points in Figure 4-5B all appear to lie
within a right triangle. From an overall data set perspective, no relationship can
be identified between fueling frequency and average fuel load. The data sets from
Add-ons G (r2 = 0.968). I (r2 = 0.837), and J (r2 = 0.616) all support the
hypothesis that fueling frequency should decrease as fuel load increases. The data
sets from Retrofits E (r2 = 0.002) and F (r2 = 0.006) do not indicate any
identifiable fueling frequency/fuel load relationship.
Low-emission Stoves. Figure 4-5C presents the fueling frequency/fuel load data for
the low-emission stove classification. The r2 values for individual stoves range
from 0.049 (Stove M) to 0.867 (Stove K). The overall r2 value is 0.473, which
indicates a fair fueling frequency/burn rate correlation.
Like the catalytic stoves, the low-emission stoves exhibit an apparent trend of
decreased fueling frequency with increased fuel load. The slope of the fueling
frequency/fuel load trend for the low-emission stoves is steeper than the slopes of
the trends indicated for all other stove technologies. This is thought to be
primarily due to the relatively smaller fireboxes in the low-emission stoves, which
would tend to limit fuel load size (all average fuel loads for the low-emission
stoves were less than 6.0 kg).
The two apparent outliers from Stove L have relatively low associated fuel loads
and conform with the expected fueling frequency/fuel load relationship, even though
they may be located outside the main data point group.
The data set from Stove L (r2 = 0.224) falls within the general trend; however, the
data points from this stove model exhibit a relatively narrow range of fuel loads
4-43
-------
(2.2 to 2.9 kg) with a relatively wide range of fueling frequencies (0.31 to 0.53
#/hr). The narrow fuel load size range could be an artifact of limitations due to
the firebox size of Stove L (37 liters), which was among the smallest fireboxes of
all low-emission stoves.
The data points for Stove M lie within the fueling frequency/fuel load trend,
although there is not a linear fueling frequency/fuel load relationship for the
Stove M data. The data points for Stove M appear to form a triangle, with data
points falling along the sides of this triangle. The ranges indicated by this
triangle are relatively small, so the nonlinear fueling frequency/fuel load
relationship for Stove M could be a statistical artifact.
Conventional Stoves. Figure 4-5D presents the fueling frequency/fuel load data for
the conventional stove classification. The conventional stoves were not separated
by stove model. The r^ value for the conventional stove data set is 0.360, which
indicates a fair fueling frequency/fuel load correlation.
The slope and data arrangement of the fueling frequency/fuel load trend for the
conventional stoves is very similar to the slope and trend location for the add-
on/retrofits. This may be expected because the add-on/retrofits are installed on
conventional stoves; therefore, firebox sizes and the fueling habits of the stove
operators may not be significantly different for the two technology
classifications. However, conventional wisdom might suggest that the trend
location should be located lower on the plot for the add-on/retrofit technology
(relatively lower fueling frequency for a given fuel load size) due to increased
efficiency from the add-on devices.
CATALYST OPERATION TIME
Catalyst Operation Time Effects on Particulate Emissions
The following discussion focuses on the relationship of catalyst operation to
particulate emissions. It is recognized that several factors in addition to
catalyst operation time affect particulate emissions; however, the objective of the
following analysis is to identify any apparent catalyst operation time/emission
rate relationships.
Conventional catalytic combustion theory suggests that as catalyst operation time
increases, particulate emissions would be expected to decrease. Because the
catalyst operation time was based on the observed temperature in the catalyst,
4-44
-------
there are artifacts of the measurement method which may complicate the anticipated
catalyst operation time/emission rate relationship. For example, an unsealed
bypass damper could result in a partially bypassed catalyst. This catalyst would
appear to be active, but may not have 100% of the flue gas stream diverted through
it, which would result in relatively higher particulate emissions.
Figures 4-6A and 4-6B are X-Y plots of catalyst operation time (%) versus
particulate emissions (g/hr) for individual sampling periods. Catalyst operation
time is defined as the percentage of time that the catalyst temperature is greater
than 260°C (SOOT) while the stove is operational (flue gas temperature greater
than 38°C [100°F]). Figure 4-6A presents the data set from the catalytic stoves
and Figure 4-6B presents the data set from the add-on/retrofits. Linear regression
coefficients (r2 values) were calculated for the individual stove models and for
the total data set in each technology category.
Catalytic Stoves. Figure 4-6A presents the catalyst operation time/emission rate
data for the catalytic stove category. The r2 values for individual stove models
range from 0.056 (Stove A) to 0.225 (Stove B). The overall r2 value is 0.045,
which indicates a very poor catalyst operation time/emission rate correlation.
The data set in Figure 4-6A does not appear to indicate any catalyst operation
time/emission rate relationship. Most data points appear to be concentrated in the
area of emission rates less than 20.0 g/hr.
Five of the six data points with emission rates greater than 25.0 g/hr are from
Stove B, which also had the highest r2 value (0.225). All six of the data points
in this area have associated catalyst operation times greater than 65%. It appears
that the relatively high emission rates associated with the higher catalyst
operation times in this area of the plot are a result of other factors with the
individual stoves; either the bypass damper or combustor is poorly sealed or the
catalyst is ineffective despite elevated temperatures.
The data set from Stove B indicates an apparent trend of increased emission rate
with increased catalyst operation time. This trend starts at approximately
(55,6.0) and continues through approximately (100,40.0). This apparent trend
contradicts conventional catalytic theory; factors other than catalyst operation
time appear to significantly affect the emission rate for Stove B.
4-45
-------
-fa
CTl
45 -
40 -
35 -
PflRTICULflTE
EMISSIONS -
(G/HR)
25 -
20 -
15 -
10 -
5 -
Figure 4-6A
Particulate Emissions (g/hr) vs. Catalyst Operation
Catalytic Stoves
oo
0
0
D
0
D
10
20
30 tO 50 60
CflTflLVST QPERflTIOH TIME (X)
80
90
ALL (R2 =0.045)
0 STDUE fl (R2 =0.056)
O STDUE B (R2 =0 .225)
D STDUE C (R* =0.129)
1 STDUE D (R2 =0.084)
X STDUE f (R2 =0.159)
-------
Figure 4-6B
Participate Emissions (g/hr) vs. Catalyst Operation
Add-On/Retrofits
35 -
PflRTICULftTE
EMISSIONS -
(G/HR)
25 -
20 -
10 -
5 -
D
D
Y
I
10
20
30
fill (R2 =0.257)
^ STDUE E (R2 =0.571)
O STDUE G (R2 =0.052)
r r i
HO 50 60
CftTflLVST DPERflTIDN TIME (X)
D STDUE F (R2 =0.333)
i STDUE J CR2 =0.708)
X STDUE I CR2 =0.190)
I
70
80
Y STDUE H
-------
The remaining data sets (Stoves A, C, D, and Stove P) do not appear to exhibit any
apparent catalyst operation time/emission rate trends. As previously mentioned,
there is a large range of catalyst operational times indicated for a given emission
rate, which makes identification of trends difficult. It can be stated that there
are no data points with catalyst operation times of less than 40% and emission
rates of less than 11.0 g/hr, as would be expected based on conventional wisdom.
Add-on/Retrofits. Figure 4-6B presents the catalyst operation time/emission rate
data for the add-on/retrofit classification. The r2 values for individual models
range from 0.062 (Add-on G) to 0.933 (Retrofit F). The overall r2 value is 0.257,
which indicates a poor catalyst operation time/emission rate correlation.
In contrast to the data set for catalytic stoves, the add-on/retrofit data set in
Figure 4-6B exhibits an apparent catalyst operation time/emission rate
relationship. If the two data points (both from Add-on I) with emission rates
greater than 24.0 g/hr and catalyst operation times greater than 50% are
disregarded, a trend of decreased emission rate with increased catalyst operation
time is apparent.
This apparent trend is developed despite the fact that the individual models of
add-on/retrofits all had relatively narrow ranges of observed catalyst operation
times, with the exception of Add-ons J and H. The ranges of catalyst operation
time (with valid emission samples) were 63.9% to 74.7% for Retrofit E, 9.8% to
25.5% for Retrofit F, 37.5% to 49.5% for Add-on G, and 53.2% to 66.6% for Add-on I.
Add-on J has a relatively wide range of associated catalyst operation times, 17.6%
to 57.8%. The data set for Add-on H consists of one data point. The relatively
narrow range of catalyst operation times for each add-on/retrofit model tends to
make the data plots for each model appear to be fairly "flat" (relatively wide
range of emission rates with a relatively narrow range of catalyst operation
times).
Catalyst Operation Time Effects on Burn Rate
Conventional catalytic combustion theory suggests that as burn rate increases,
catalyst operation time would be expected to increase also. At high burn rates
high flue gas temperatures would be expected to maintain catalyst lightoff; lower
flue gas temperatures at lower burn rates may tend to make the catalyst become
inactive, which would be reflected by lower catalyst operation times.
4-48
-------
As in the case of the catalyst operation time/emission rate discussion, an unsealed
bypass damper or failed catalyst may complicate the catalyst operation time/burn
rate relationship. If a large percentage of the flue gas is diverted around the
catalyst, it may not receive sufficient fuel to maintain lightoff. A failed
catalyst could result in lower efficiency, so relatively higher burn rates would be
required to maintain desired heat output. This situation would result in
relatively high burn rates with relatively high apparent catalyst operation time.
Figures 4-7A and 4-7B are plots of catalyst operation time (%) versus burn rate
(kg/hr) for individual sampling periods. Figure 4-7A presents the data set from
the catalytic stoves and Figure 4-7B presents the data set from the add-on/
retrofits. Linear regression coefficients (r2 values) were calculated for the
individual stove models and for the total data set in each technology category.
Catalytic Stoves. Figure 4-7A presents the catalyst operation time/burn rate data
for the catalytic stove classification. The r2 values for individual models range
from 0.003 (Stove B) to 0.743 (Stove A). The overall r2 value is 0.250, which
indicates a poor catalyst operation time/burn rate correlation.
The overall data set in Figure 4-7A appears to reflect the expected relationship of
increased catalyst operation time with increased burn rate, although there is
considerable variability.
The data sets from Stove A (r2 = 0.743) and Stove C (r2 = 0.442) appear to exhibit
fairly well-defined catalyst operation time/burn rate trends as would be expected.
The data sets from Stove B (r2 = 0.003), Stove D (r2 = 0.077), and Stove Code P (r2
= 0.026) do not appear to exhibit catalyst operation time/burn rate relationships
as would be expected. The data sets from each of these stoves contain individual
data points that are located along the overall trend; however, conclusive
identification of catalyst operation time/burn rate relationships for these
individual stove models is difficult.
The range of catalyst operation time for 11 of the 13 data points for Stove B is
67.3% to 90.9%. The remaining two data points have catalyst operation times of
45.0% and 42.9%. The burn rate range for Stove B is 0.84 to 1.57 kg/hr. This is
an indication that for the majority of the Stove B data set (11 of 13 data points)
the catalyst operation time remained relatively high over a wide range of burn
rates. This characteristic would be desirable in a catalytic stove; however, as
4-49
-------
2.5 -
2.0 -
BURN
RflTE
(KG/HR)
c_n
CD
1.5-
1.0 -
0.5 -
Figure 4-7A
Burn Rate (kg/hr) vs. Catalyst Operation (X)
Catalytic Stoves
X D
D D
:
°
10
20
30 tO 50 SO
CftTflLVST QPERflTIDN TIME t.y.1
70
80
30
ALL (R2 -Ci.250)
0 STDUE fl (R2 =0.74-3)
0 STDUE B CR2 =0.003)
D SIDUE C (ft2 =0.tt2)
§ STDUE D ( R2 =0.077)
X STDUE f ( R2 =0.026)
-------
2.5 -
2.0 -
BURN
RATE
(KG/HR)
1.5 -
1.0 -
0.5 -
Figure 4-7B
Burn Rate (kg/hr) vs. Catalyst Operation (x)
Add-On/Retrofits
* X
X
D
o •
0 Q •
0 0
0
D
10
20
I
30
ftLL (R2 =0.238)
0 STDUE E CR2 =0.001)
O STDUE G (R2 =0.781)
I I I
HO 50 SO
CftTftLVST QPERflTIDN TIHE (X)
D STDUE F (R2 =0.628)
| STDUE J CR2 =0.808)
X STDUE I CR2 =0.002)
I
70
\
80
I
90
Y STDUE H (R2 = Hfl )
-------
mentioned in the catalyst operation time/emission rate analysis section, Stove B
appeared to have increased emissions with increased catalyst operation time.
The data points from Stove D appear to fit in the overall catalyst operation
time/burn rate trend; however, the r2 value for Stove D (0.077) indicates a poor
correlation. This may be an artifact of the relatively narrow range of burn rates
observed for Stove D (0.58 to 1.27 kg/hr). This narrow range of burn rates tends
to make the slope of the catalyst operation time/burn rate curve steeper in
relation to the catalytic stove models where a wider range of burn rates was
observed. The majority of data points for Stove D appear to be located in a
rectangle bounded by burn rates of approximately 0.55 to 1.25 and by catalyst
operation times of approximately 32% to 85%.
The data points from Stove P also do not exhibit a clearly defined catalyst
operation time/burn rate relationship. Five of the six data points are in a
relatively narrow range of burn rates (1.01 to 1.18 kg/hr) accompanied by a
relatively wide range of catalyst operation times (33.8% to 87.8%). These points
appear to fit in the overall trend; however, the data points at either end of the
range of catalyst operation time are on the fringe of the data set. The sixth data
point from Stove P, (58.8,1.83), appears to be an outlier.
Add-on/Retrofits. Figure 4-7B presents the catalyst operation time/burn rate data
for the add-on/retrofit classification. The r2 values for individual models range
from 0.001 (Retrofit E) to 0.808 (Add-on J). The overall r2 value is 0.238, which
indicates a poor catalyst operation time/burn rate correlation.
Although the r2 value for the add-on/retrofit classification is similar to the r2
value for the catalytic stove classification (0.288), the add-on/retrofit plot in
Figure 4-7B does not appear to exhibit an identifiable catalyst operation time/burn
rate correlation as did Figure 4-7A.
Add-on J (r2 = 0.808) is the only model in the add-on/retrofit classification where
the data points appear to follow the expected catalyst operation time/burn rate
correlation. The five data points in this data set exhibit a linear trend.
The data sets from Retrofit E (r2 = 0.001) and Add-on I (r2 = 0.002) do not appear
to exhibit any catalyst operation time/burn rate relationship; however, this is
probably an artifact of the narrow data ranges for each of these devices. It
4-52
-------
appears that the individual data points for each of these devices are so closely
grouped that no catalyst operation time/burn rate trends are evident.
The data set from Retrofit F may appear to exhibit an increase in catalyst
operation time with an increase in burn rate; however, this interpretation should
be made with caution due to the small data set (three points) and the relatively
wide range of burn rates observed (0.97 to 1.59 kg/hr) in conjunction with a
relatively narrow range of catalyst operation times (9.8% to 25.5%). The apparent
catalyst operation time/burn rate trend is predominantly indicated by one data
point, (25.5,1.59).
The four-point data set from Add-on G (r^ = 0.781) indicates a catalyst operation
time/burn rate relationship that contradicts conventional wisdom. As in the case
of Retrofit F, this trend is indicated based on a single data point, (9.4,1.86), so
caution should be used in considering this apparent trend. The four data points
for Add-on G are grouped in a relatively narrow range of burn rates (1.61 to 1.86
kg/hr) and have a relatively wide range of associated catalyst operation times
(9.4% to ,49.5%). If the data point at (9.4,1.86) is considered an outlier, no
catalyst operation time/burn rate relationship is indicated due to the remaining
narrow range of burn rates (1.61 to 1.70 kg/hr) and catalyst operation times (37.5%
to 49.5%).
Catalyst Operation Time Effects on Creosote Accumulation
Conventional catalytic combustion theory suggests that as catalyst operation time
increases, creosote accumulation would be expected to decrease due to a lower
particulate concentrations in the flue gas stream.
Caution should be used when interpreting the creosote accumulation data due to
uncertainties associated with the creosote accumulation measurements. For example,
creosote could be volatilized during periods of high flue gas temperatures. In
addition, factors other than catalyst operation (chimney system type, particulate
emission rate, and burn rate) can significantly affect creosote accumulation.
Figures 4-8A and 4-8B are X-Y plots of overall mean catalyst operation time (%)
versus creosote accumulation (kg/1000 HDD) for individual homes. The overall mean
catalyst operation time (%) was calculated for each home and plotted against the
overall creosote accumulation. Figure 4-8A presents the data set from the
catalytic stoves and Figure 4-8B presents the data set from the add-on/retrofits.
4-53
-------
Figure 4-8A
Creosote Accumulation (kg/1888 HDD) vs. Catalyst Operation (x)
Catalytic Stoves
2.5 -
2.0-
CREQSDTE
flCCUMULRTIDN
(KS/iOOO HDD)
1.5-
1.0 -
0.5 -
D
I
D
I
D
o
X
10
20
30 tO 50 60
CflTflLVST DPERflTIDN TIHE (X)
70
80
90
ALL (R2 =0,033)
0 STDUE ft (R2 = Hfl )
O STDUE B (R2 =0.257)
D S1DUE C (R2 =0,554)
I STDUE D (R2 =0.89t)
X STDUE P (R2 =0.535)
-------
Figure 4-8B
Creosote Accumulation (Jigr/1880 HDD) vs. Catalyst Operation (x)
Add-On/Retrofits
I
t-n
en
2.5 -
2.0 -
CREDSDTE
ftCCUHULfiTIDN
(KG/1000 HDD)
1.5 -
1.0 -
0.5 -
Y
I
X
0 °
. Y X 0
1 Y
Y
I 1 1 1 1 1 1 1 1
10 20 30 YO SO 60 70 80 30
CftTftLVST DPERftTIDN TIME C/.1
flLL (R2 =0.001) D STDUE G (R2 = Nfl ) Y STDUE J (R2 =0.018)
0 STDUE E (R2 = Hfl ) I STDUE H CR2 = Nfi )
0 STDUE F (R2 = Hft ) X STDUE I CR2 = Hfi )
-------
Linear regression coefficients (r2 values) were calculated for the individual stove
models and for the total data set in each technology category.
Catalytic Stoves. Figure 4-8A presents the catalyst operation time/creosote
accumulation data for the catalytic stove classification. The r2 values for
individual models range from 0.257 (Stove B) to 0.894 (Stove D). The overall r2
value is 0.039, which indicates a very poor catalyst operation time/creosote
accumulation correlation.
The overall data set for the catalytic stove classification does not appear to
exhibit a catalyst operation time/creosote accumulation relationship; the
individual points in the overall data set appear to be randomly located.
The data sets for Stoves A (r2 not calculated, two data points), B (r2 = 0.257), C
(r2 = 0.554), and D (r2 = 0.894) all appear to exhibit individual trends of
increased creosote accumulation with increased catalyst operation, which
contradicts conventional wisdom. These apparent trends should be interpreted with
caution due to the chimney type and operating practice factors that can
significantly influence creosote accumulations in individual homes.
The data set from Stove A consists of two data points from two different homes.
Home N01 had the lowest overall catalyst operation time (26.0%) and the lowest
creosote accumulation (0.61 kg/1000 HDD). Home N10 had a higher catalyst operation
time (86.3%) and a higher creosote accumulation (1.15 kg/1000/HDD). Home N01 had a
prefabricated metal chimney, so the creosote accumulation would be expected to be
lower than that in Home N10, which had an exterior masonry chimney. The overall
mean burn rate in Home N10 (1.42 kg/hr) was twice as high as the overall mean burn
rate in Home N01 (0.70 kg/hr). Higher catalyst operation time might therefore be
anticipated in Home N10 along with higher creosote accumulation (because of the
exterior masonry chimney).
The data set from Stove B also exhibited an apparent trend of increased creosote
accumulation with increased catalyst operation time. However, this trend is
indicated by a single data point, (54.1,0.44). This data point is from Home Vll,
which had an interior masonry chimney (the three other homes using Stove B had
exterior masonry chimneys). As in the case of Stove A, the home with the interior
chimney system had relatively lower creosote accumulation accompanied by a
relatively lower catalyst operation time. If the data point from Home Vll is
disregarded, the remaining three data points exhibit a creosote accumulation range
4-56
-------
| of 0.54 to 1.13 kg/1000 HDD with a catalyst operation time range of 77.2% to 90.9%.
: This is a relatively narrow range of catalyst operation time accompanied by a
relatively wide range of creosote accumulation, indicating that no clearly defined
relationship between catalyst operation and creosote accumulation was found in
these three homes.
The data set from Stove C also exhibited an apparent trend of increased creosote
accumulation with increased catalyst operation time. As in the case of Stove B,
this trend is indicated by a single data ppint. This data point is from Home N03,
which had an interior masonry chimney (the other two homes using Stove C had
exterior masonry chimneys). As in the case of Stoves A and B, the home with the
interior chimney system had relatively lower creosote accumulation accompanied by a
relatively lower catalyst operation time. Home N03 also had the lowest overall
mean burn rate (1.00 kg/hr) for Stove C (Homes V07 and V16 had overall mean burn
rates of 1.60 and 1.21 kg/hr, respectively). As in the case of the data set from
Stove A, higher catalyst operation times would be anticipated in Homes V07 and V16
along with higher creosote accumulations because of the exterior masonry chimneys.
The data set from Stove D also exhibits an apparent trend of increased creosote
accumulation with increased catalyst operation time. This trend is fairly well
defined by four data points (r2 = 0.894). All of the homes using Stove D had
exterior masonry chimney systems. All creosote accumulations for Stove D are
relatively low (range of 0.35 to 0.73 kg/1000 HDD). Consequently, the
interpretation of this apparent trend should be made with caution due to
uncertainties associated with the creosote accumulation measurements.
Each of the four catalytic stove models evaluated showed an apparent trend of
increased creosote deposition with increased catalyst operation time. This
observation contradicts conventional wisdom; creosote accumulation would be
expected to decrease as catalyst activity increases. Caution should be used in
interpreting these trends due to uncertainties associated with the creosote
11 measurement methods, the small sample sizes, and the demonstrated influence of
chimney system type. A more detailed analysis and a larger data set is recommended
before drawing conclusions regarding the relationship of catalyst operation time to
ncreosote accumulation for catalytic stoves.
H.
II
Add-on/Retrofits. Figure 4-8B presents the catalyst operation time/creosote
accumulation data for the add-on/retrofit classification. An r2 value was
calculated only for Add-on J (r2 = 0.018) because this was the only data set with
4-57
-------
more than two data points. The overall r2 value is 0.001, which indicates a very
poor catalyst operation time/creosote accumulation correlation.
As previously discussed, this technology classification had a high number of
exterior masonry chimney systems relative to the other three technology
classifications, which probably resulted in a high bias of creosote accumulation
values.
Caution should be used when interpreting the overall catalyst operation
time/creosote accumulation trend indicated for the add-on/retrofits. This trend is
based on only ten data points, has three apparent outliers, and is represented in a
relatively low range of creosote accumulations (0.20 to 0.79 kg/1000 HDD). As in
the case of the catalytic stoves, further detailed analysis and a larger data set
are recommended.
ALTERNATE HEATING SYSTEM EFFECTS
Alternate Heating System Effects on Particulate Emissions
The following discussion of alternate heating system usage was separated into three
frequency-of-use categories. The first category (which usually contains the
majority of data points) consists of alternate heat usage of 0.0%. The second
category consists of alternate heat usage in the 0.1% to 3.5% range. This range of
alternate heat use approximately corresponds to alternate heat usage of up to 50
minutes per day, and probably reflects the situation where the alternate heating
system is used to briefly supplement the woodstove heat output when the stove is
being "fired up." The third category consists of alternate heat usage greater than
3.5%, which can be indicative of frequent alternate heating system use in
conjunction with woodstove operation.
It is difficult to hypothesize an alternate heat use/particulate emissions
relationship. Many factors can influence the amount of observed alternate heat
usage in individual homes, including woodstove heat output, lifestyle of the
homeowners, usage pattern of the alternate heat source, ability of the home to
retain heat, indoor temperature desired by the homeowners, and homeowners' desire
to use only wood fuel. Similarly, as documented in previous sections, many factors
can affect particulate emissions.
4-58
-------
It should be noted that only heat ducts in the room with the woodstove were
monitored; a central heating system could be heating other rooms in a home not
conveniently served by a woodstove.
Figures 4-9A through 4-9D are plots of alternate heating system use (%) versus
particulate emissions (g/hr) for individual sampling periods for the four stove
technology types evaluated (catalytic stoves, add-on/retrofits, low-emission
stoves, and conventional stoves). Alternate heating system use is defined as the
percentage of time while the stove is operational (flue gas temperature greater
than 38°C [100°F]) that the alternate heat system is in use (as indicated by a
thermal sensor placed in the furnace duct or electric baseboard). In homes where
zoned electric baseboard heat was used, this alternate heating system percentage
may be indicative of alternate heat usage only in the room with the stove. Linear
regression coefficients (r2 values) were calculated for the individual stove models
and for the total data set in each technology category.
Catalytic Stoves. Figure 4-9A presents the heating system use/emission rate data
for the catalytic stove classification. The r2 values for individual stove models
range from 0.000 (Stove D) to 0.225 (Stove A). The overall r2 value is 0.024,
which indicates a very poor heating system use/emission rate correlation.
Twenty-two of the 59 data points (37% of the data set) for the catalytic stoves
have a heating system use above 0.0%. Of these points, nine (15% of the data set)
have a heating systems use above 3.5%. These nine points are from three homes
(V07, Stove C, four points; V16, Stove C, three points, and Nil, Stove D, two
points). This indicates that for the overall catalytic stove data set that the
majority of data points (63%) represent 0.0% heat use, 22% had heat use values in
the 0.1% to 3.5% range, and 15% (from three homes) had heat use values above 3.5%.
Of the three homes with heating system use greater than 3.5%, participants in two
(V07 and Nil) expressed dissatisfaction with the heat output of their stoves (refer
to Appendix A, Table A-2). Participants in Home V16 were pleased with the
performance of Stove C. None of the remaining homeowners who used the catalytic
stoves (all with heating system use less than 3.5%) expressed dissatisfaction with
the heat output of the catalytic stoves.
Because the majority of data points have heating system use percentages of less
than 3.5%, it is difficult to identify a heating system use/emission rate
relationship for the catalytic stoves.
4-59
-------
Figure 4-9A
Participate Emissions (g/hr) vs. Heating System Use (X)
Catalytic Stoves
35 -
PflRTICULflTE
EMISSIONS -
25 -
i
CTl
o
D
D
D
D
D
10
D
D
D
1
5
I
10
1
15
i
20
1
25
1
30
1
35
1
fO
HEflTING SYSTEM USE (X)
FILL (R2 =0.021)
0 STDUE fl (R2 =0 .225)
O STDUE B (R2 =0 .180)
D STDUE C (R2 =0.014)
| STDUE 0 (R2 =0.001)
X STDUE P (R2 = Hfl )
-------
t5 -
fO -
I
CT>
35 H
PfiRTICULflTE
EMISSIONS '
(G/HR)
25 •
20 -
IE
if
Figure 4-9B
Participate Emissions (g/hr) vs. Heating System Use (x)
Add-On/Retrofits
D
Y
I
10
I i i
15 20 25
HEflTING SVSTEM USE (X)
1
30
35
fO
ftLL (R2 =0,02S)
0 STDUE E (R2 =0,259)
O STDUE G (R2 =0.750)
D STDUE F (R2 =0.15*)
| STDUE J (R2 =0.081)
X STDUE I CR2 : Nfl )
STDUE H (R2 = Hfi )
-------
Figure 4-9C
Particulate Emissions (g/hr) vs. Heating System Use (x)
Low-Emission Stoves
cr>
ro
10 -
35 -
PflRTICULflTE
EMISSIONS '
(G/HR)
25 -
20 -
15 -
10
C —
0
o
CP
0
10
ALL (R2 =O
0 STDUE N (R2 =0.010)
O STDUE L (R2 =0 .005)
15 20 25
HEfiTING SVSTEM USE (X)
30 35
D STDUE H (R2 =0,304)
§ STDUE K (R2 =0 .010)
fO
-------
35
PflRTICULflTE
EMISSIONS
(G/HR3
25
20
IE
10
Figure 4-9D
Participate Emissions (g/hr) vs. Heating System Use ('/)
Conventional Stoves
10
I
20
I
25
I
30
I
35
I
to
HEftTINQ SYSTEM USE
-------
Add-on/Retrofits. Figure 4-9B presents the heating system use/emission rate data
for the add-on/retrofit classification. The r2 values for individual_ models range
from 0.084 (Add-on J) to 0.750 (Add-on G). The overall r2 value is 0.025, which
indicates a very poor heating system use/emission rate correlation.
As in the case of the catalytic stoves, it is difficult to identify any heating
system use/emission rate trend from the data set in Figure 4-9B. All of the 19
data points have a heating system use of less than 5.6%. Nine points (47% of the
data set) have a heating system use of 0.0%. Three points (16% of the data set,
all from Home V10) have heating system usage in the range of 3.5% to 5.5%.
Add-on G has an r2 value that indicates a good heating system use/emission rate
correlation (0.750); however, this should be viewed with caution. The data set for
Add-on G consists of three data points, with only one alternate heat use value
greater than 0.0%. The high r2 value is probably not an indication of a heating
system use/emission rate relationship.
Low-emission Stoves. Figure 4-9C presents the heating system use/emission rate
data for the low-emission stove category. The r2 values for individual stove
models range from 0.005 (Stove L) to 0.304 (Stove M). The overall r2 value is
0.006, which indicates a very poor heating system use/emission rate correlation.
The low-emission stoves had the highest percentage of data points (45% of the data
set) with heating system use in the 0.1% to 3.5% range of all stove technologies.
This may be a result of the low-emission stove designs (i.e., relatively smaller
firebox size), which generally dictate shorter burn times relative to the other
stove technologies evaluated. Although the data set from the low-emission stoves
does not include a significant number of heating system use values greater than
3.5%, the relatively higher percentage of heating system use values in the 0.1% to
3.5% range may be an indication of homeowners using their alternate heat systems to
supplement their woodstoves.
Even though a relatively large percentage of homes where the low-emission stoves
were evaluated also used some amount of supplemental alternate heat, no homeowners
reported dissatisfaction with heat output of the low-emission stoves.
It is difficult to identify any heating system use/emission rate correlation from
the low-emission stove data set in Figure 4-9C. Of the 24 data points, ten (42% of
the data set ) have a heating system use of 0.0%. Three data points (13% of the
4-64
-------
data set) have heating system usage in the range of 3.5% to 11.1% (11.1% is the
highest heating system use in the low-emission stove data set).
Conventional Stoves. Figure 4-9D presents the heating system use/emission rate
data for the conventional stove classification. The conventional stoves were not
separated by stove model. The overall r2 value is 0.015, which indicates a very
poor heating system use/emission rate correlation.
Three of the 14 data points (21% of the data set) for the conventional stove in
Figure 4-9D have heating system use values greater than 0.0%. These three data
points have heating system use values of 3.3%, 3.1%, and 3.8%. With the majority
of data points (79% of the data set) for the conventional stoves having heating
system use values of 0.0%, it is difficult to identify a heating system
use/emission rate relationship.
Alternate Heating System Effects on Burn Rate
A conventional hypothesis suggests that if burn rate is considered to be roughly
proportional to heat output, alternate heating system use may be proportional to
burn rate. At low burn rates (low stove heat output) a relatively higher
percentage of heating system use would be anticipated. At higher burn rates (high
stove heat output) a relatively lower percentage of heating system use would be
anticipated. This theory is significantly complicated by the factors mentioned
previously which influence the percentage of alternate heat usage.
Figures 4-1OA through 4-10D are X-Y plots of alternate heating system use (%)
versus burn rate (kg/hr) for individual sampling periods for the four stove
technology types evaluated (catalytic stoves, add-on/retrofits, low-emission
stoves, and conventional stoves). Linear regression coefficients (r2 values) were
calculated for the individual stove models and for the total data set in each
technology category.
Catalytic Stoves. Figure 4-1OA presents the heating system use/burn rate data for
the catalytic stove classification. The r2 values for individual stove models
range from 0.047 (Stove B) to 0.401 (Stove C). The overall r2 value is 0.026,
which indicates a very poor heating system use/burn rate correlation.
As in the case of the heating system use/emission rate plot, the majority of data
points (41 of 67, 61%) have a heating system use of 0.0%. Of the remaining 26 data
points, 14 (21% of the data set) have heating system usage in the range of 0.1% to
4-65
-------
BURN
RflTE
(KG/HR)
en
cr\
1 .5
1 .0
Figure 4-18A
Burn Rate (kg/hr) vs. Heating System Use (x)
Catalytic Stoves
2.5 -
D
D
D
D
D
0.5 -
10
15 20 25
HEflTING SYSTEM USE (X)
ftLL (R2 =0.025)
Q STDUE ft (R2 =0.087)
O STDUE B (R2 =0 ,0t7)
30 35
D StOUE C (R2 =0 tOl)
I STDUE D (R2 =0,12t)
X STDUE P (R2 = Hfl )
fO
-------
2.5
BURN
RflTE
(KG/-HR)
Figure 4-1BB
Burn Rate (kg/hr) vs. Heating System Use (x)
Add-Qn/Retrofits
I
CT,
1.5
1.0
0.5
•Y
10
ALL
-------
BURN
RflTE
(KG/HR)
en
Co
Figure 4-1BC
Burn Rate (kg/hr) vs. Heating System Use
Low-Emission Stoves
2.5 -
1.5 -
1.0 -}
0.5 -
Eb
D Q
10
ftLL (R2 =0,lf4)
0 STDUE N (R2 =O.SfO)
O STDUE L (R2 =0 .327)
15 20 25
HEftTING SVSTEH USE (x)
30 35
D STDUE H (ft2 =0.027)
• STDUE K (R2 =0.085)
-------
O1
UD
2.5
BURN
RflTE
(KG/HR)
i.S
1.0
0.5
Figure 4-10D
Burn Rate (kg/hr) vs. Heating System Use (x)
Conventional Stoves
15 20
HEftTIHG SVSTEH USE (X)
I
25
-------
3.5%, and 12 (18% of the data set) have heating system use greater than 3.5%. The
12 data points with heating system use greater than 3.5% are from three homes, V07
(Stove C), V16 (Stove C), and Nil (Stove D). As previously mentioned, participants
in Homes V07 and Nil expressed dissatisfaction with the heat output performance of
the catalytic stoves.
It is difficult to identify any heating system use/burn rate trends from the
overall data set or from the data sets for individual catalytic stove models.
The best heating system use/burn rate correlation appears to be for Stove C (r2 =
0.401). The range of heating system use for this stove model is relatively wide
(0.0% to 39.1%). The burn rate range is 0.97 to 1.89 kg/hr; however, there is only
one data point with a burn rate greater than 1.50 kg/hr. This may tend to give a
visual sense of increased heating system use with increased burn rate when the data
set is viewed as a whole; however, the remaining points for Stove C do not appear
to conclusively exhibit any identifiable heating system use/burn rate trend. The
data set for Stove C may be biased because two of the three homes where the stove
was evaluated had relatively high levels of alternate heat use.
Add-on/Retrofits. Figure 4-10B presents the heating system use/burn rate data for
the add-on/retrofit classification. The r2 values for individual models range from
0.002 (Add-on G) to 0.262 (Retrofit F). The overall r2 value is 0.160, which
indicates a poor heating system use/burn rate correlation.
As in the case of the heating system use/emission rate analysis, there is a
relatively small percentage of data points in this technology classification that
have a heating system use greater than 3.5%. Of the 24 data points, 13 (54% of the
data set) have a heating system use of 0.0%, eight (33% of the data set) have a
heating system use in the range of 0.1% to 3.5%, and three (13% of the data set,
all from Home V10) have heating system usage values greater than 3.5% (maximum
heating use 5.5%). The predominance of low heating system use values in this
technology classification makes identification of a heating system use/burn rate
relationship difficult.
Low-emission Stoves. Figure 4-10C presents the heating system use/burn rate data
for the low-emission stove classification. The r2 values for individual stove
models range from 0.027 (Stove M) to 0.840 (Stove N). The overall r2 value is
0.144, which indicates a poor heating system use/burn rate correlation.
4-70
-------
Of the 26 data points, 12 (46% of the data set) have a heating system use of 0.0%,
11 (42% of the data set) have heating system usage in the range of 0.1% to 3.5%,
and three (12% of the data set) have heating system usage values greater than 3.5%
(maximum heating use 11.1%). As in the case of the heating system use/emission
rate analysis, the low-emission stoves have the highest percentage of data points
with heating system usage range of 0.0% to 3.5%.
The data set from Stove N appears to show a good heating system use/burn rate
correlation (r^ = 0.840). However, caution should be used in making this
interpretation. Three of the five data points from Stove N have associated heating
system use values of 0.0% or 0.1%. The remaining two data points have heating
system use values of 1.8% and 2.5%. Because of the small data set and the lack of
any heating system use values above 2.5%, the relatively high r2 value for the
Stove N data set may be a statistical coincidence rather than an indication of a
heating system use/burn rate correlation.
The overall data set for the low-emission stoves does not indicate any identifiable
heating system use/burn rate relationship. There are not sufficient data points
with heating system use values above 3.5% to make any heating system use/burn rate
relationship evident.
Conventional Stoves. Figure 4^100 presents the heating system use/burn rate data
for the conventional stove classification. The conventional stoves were not
separated by stove model. The overall r^ value is 0.043, which indicates a poor
heating system use/burn rate correlation.
Of the 16 data points in Figure 4-10D, 12 (75% of the data set) have heating system
use values of 0.0%, three (19% of the data set) have heating system usage in the
0.1% to 3.5% range, and one (6% of the data set) has a heating system use value
above 3.5% (the heating system use value for this data point is 3.8%).
Because the majority of data points for the conventional stoves have heating system
use values that are 0.0%, it is difficult to identify a heating system use/burn
rate relationship from this data set.
CHIMNEY SYSTEM EFFECTS
Table 4-1 presents data on the creosote accumulation (kg/1000 HDD), emission rate
(g/hr), and burn rate (kg/hr) for each stove technology type by chimney system
type. The data are separated into four general categories of chimney systems.
4-71
-------
Table 4-1
CHIMNEY SYSTEM EFFECTS ON CREOSOTE ACCUMULATION, EMISSION RATE, AND BURN RATE
Technology
Catalytic
Stoves
Add-on/
Retrofits
Low-
Emission
Stoves
Conven-
tional
Stoves
Chimney
Typeb/
a
b
c
d
a
b
c
d
a
b
c
d
a
b
c
d
Creosote Accumulation3/
(kg/1000 HDD)
X
0.48
0.64
0.73
0.50
0.30
0.59
1.24
0
0.24
0.52
0.69
1.10
0.62
0.86
1.63
0.80
a
0.36
0.43
0.37
0.32
0.15
0.14
0.99
0
0.07
0.35
0.36
0.03
0.23
0.46
1.33
0.20
N
11
14
18
5
4
4
17
0
5
6
2
2
7
11
13
4
Range
0.04-1.14
0.06-1.98
0.13-1.43
0.06-0.93
0.14-0.52
0.36-0.73
0.33-3.59
0.15-0.36
0.11-1.08
0.33-1.05
1.07-1.13
0.18-0.86
0.07-1.80
0.33-5.78
0.52-1.02
Emission Rate3'
(g/hr)
X
18.2
13.6
17.8
17.6
7.3
25.9
16.3
0
16.4
8.8
13.6
11.2
0
10.6
25.3
0
a
3.1
4.9
10.0
11.7
0
8.6
7.8
0
12.9
5.6
7.5
1.8
0
5.8
6.7
0
N
6
20
28
4
1
4
14
0
10
6
5
2
0
5
9
0
Range
13.0-21.9
5.5-24.3
1.7-41.3
6.3-34.6
16.5-36.7
6.3-37.3
3.6-47.6
2.0-18.3
4.3-26.3
9.4-12.9
2.9-17.3
13.9-34.0
Burn Rate3/
(kg/hr)
X
0.85
1.02
1.30
1.61
1.31
1.15
1.69
0
1.06
0.98
1.01
0.86
0
1.69
1.76
0
CT
0.20
0.19
0.26
0.46
0
0.28
0.41
0
0.17
0.25
0.12
0.03
0
0.40
0.41
0
N
8
23
31
4
1
5
18
0
10
7
6
3
0
6
10
0
Range
0.57-1.18
0.61-1.27
0.84-1.89
1.01-2.26
0.87-1.59
1.01-2.35
0.76-1.34
0.67-1.38
0.85-1.18
0.84-0.90
1.12-2.45
0.92-2.45
-p.
I
a/ For each chimney type applicable to each parameter, the mean (x), standard deviation (a)
and range of values is presented.
sample population (N),
Chimney systems are classified as follows:
a. Round prefabricated metal chimneys with six-inch, seven-inch, or eight-inch inside diameters (chimney types I,
II, III, and XI in home characteristics table).
b. Rectangular tile-lined masonry chimneys located inside the exterior walls of the house with 7" x 7" or 7" x 11
flue cross-section sizes (chimney types V and VI in home characteristics table).
c. Rectangular tile-lined masonry chimneys located outside the exterior walls of the house with 7" x 7" or 7" x
11' flue cross-section sizes (chimney types VII, VIII, and IX in home characteristics table).
d. Chimneys that do not fit into above categories a, b, or c (chimney types IV, X, XII, and XIII in home
characteristics table).
-------
For each value reported by chimney system, the mean, standard deviation, sample
population, and range of values are presented.
The chimney systems were categorized into four groups. The fourth group contains
the chimney types that did not fit into three basic categories. The chimney system
categories include:
a. Round prefabricated metal chimneys with 15 cm (6"), 18 cm (7"), or
20 cm (8") inside diameters.
b. Rectangular tile-lined masonry chimneys located inside the exterior
walls of the house with 18 cm by 18 cm (7" by 7") or 18 cm by 28 cm
(7" by 11") flue cross-section sizes.
c. Rectangular tile-lined masonry chimneys located outside the exterior
walls of the house with 18 cm by 18 cm (7" by 7") or 18 cm by 28 cm
(7" by 11") flue cross-section sizes.
d. Chimney systems not defined by a, b, or c above.
The data for the fourth category ("d") is presented in Table 4-1; however, the
chimney types in this category have significantly different construction features,
so any calculated means for this category should not be compared with the means of
the other three chimney system types. The chimney types in category "d" include
round tile-lined masonry chimneys, stainless steel-lined masonry chimneys, and
stoves vented into conventional masonry fireplaces.
For the Group I, II, and III homes (used for the creosote accumulation analysis),
the mixture of chimney type by technology category was as follows:
• Catalytic Stoves: Seven (23%) prefabricated metal, 9 (29%) interior
masonry, 11 (35%) exterior masonry, and four (13%) other.
• Add-on/Retrofits: Three (16%) prefabricated metal, four (21%)
interior masonry, and 12 (63%) exterior masonry.
• Low-emission Stoves: Four (33%) prefabricated metal, five (42%)
interior masonry, two (17%) exterior masonry, and one (8%) other.
• Conventional Stoves: Six (20%) prefabricated metal, ten (33%)
interior masonry, 11 (37%) exterior masonry, and three (10%) other.
For the Group I and Group III (instrumented) homes used for the emission rate and
burn rate analysis, the mixture of chimney type by technology category was as
follows:
• Catalytic Stoves: Three (17%) prefabricated metal, six (33%)
interior masonry, seven (39%) exterior masonry, and two (11%) other.
4-73
-------
• Add-on/Retrofits: one (11%) prefabricated metal, two (22%) interior
masonry, and six (67%) exterior masonry.
• Low-emission Stoves: Four (40%) prefabricated metal, three (30%)
interior masonry, two (20%) exterior masonry, and one (10%) other.
• Conventional Stoves: Two (33%) interior masonry, and four (67%)
exterior masonry.
Chimney System Effects on Creosote Accumulation
As previously discussed in Section 3 of this report, the chimney system
construction and location (interior or exterior) appear to have a significant
effect on creosote accumulation. In general, it appears that the prefabricated
metal chimneys have the lowest creosote accumulations, interior masonry chimneys
have mid-range creosote accumulations, and exterior masonry chimneys have the
highest creosote accumulations. Caution should be used in interpreting the
creosote accumulation data presented in Table 4-1, as there may be several
different stove models (each of which has unique design characteristics)
represented for a given stove technology classification and chimney type. Also,
inherent difficulties with creosote accumulation measurements (as previously
discussed) may significantly affect the mean accumulations presented in Table 4-1.
Catalytic Stoves. The mean creosote accumulations by chimney type in the catalytic
stove classification are ranked as would be anticipated. The prefabricated metal
chimneys had the lowest mean creosote accumulation (0.48 kg/1000 HDD). The
interior masonry chimneys had the middle mean creosote accumulation (0.64 kg/1000
HDD). The exterior masonry chimneys had the highest mean creosote accumulation
(0.73 kg/1000 HDD).
The data set for the catalytic stoves had nine or more data points for each chimney
system type. Caution should be used in this interpretation, however. The range of
overall mean creosote accumulations is relatively small (0.48 to 0.73 kg/1000 HDD).
When the standard deviations for the mean creosote accumulation for each chimney
type are considered, there is potential overlap of the data sets for the chimney
types, which may indicate that the differences in mean creosote accumulations may
be a statistical artifact.
Add-on/Retrofits. As in the case of the catalytic stoves, the mean creosote
accumulations by chimney type in the add-on/retrofit classification are ranked as
would be anticipated. The prefabricated metal chimneys had the lowest mean
creosote accumulation (0.30 kg/1000 HDD). The interior masonry chimneys had the
4-74
-------
middle mean creosote accumulation (0.59 kg/1000 HDD). The exterior masonry
chimneys had the highest mean creosote accumulation (1.24 kg/1000 HDD).
As in the case of the catalytic stoves, if the standard deviations for each
creosote accumulation mean are considered, there is potential overlap of the data
sets for each chimney type. However, the magnitude of the means increases by a
factor of approximately two from the prefabricated metal chimney mean to the
interior masonry chimney mean, and by a factor of approximately two from the
interior masonry chimney mean to the exterior masonry chimney mean. This is an
indication that a fairly high degree of confidence can be placed in the relative
ranking of creosote accumulation by chimney type for the add-on/retrofit data set.
Low-emission Stoves. As in the case of the catalytic stoves and add-on/retrofits,
the mean creosote accumulations by chimney type in the low-emission stove
classification are ranked as would be anticipated. The prefabricated metal
chimneys had the lowest mean creosote accumulation (0.24 kg/1000 HDD). The
interior masonry chimneys had the middle mean creosote accumulation (0.52 kg/1000
HDD). The exterior masonry chimneys had the highest mean creosote accumulation
(0.69 kg/1000 HDD).
As in the case of the add-on/retrofits, the mean creosote accumulation for the
interior masonry chimneys was higher by a factor of two than the mean creosote
accumulation for the prefabricated metal chimneys. There is a relatively small
difference (0.17 kg/1000 HDD) in the mean creosote accumulations for the interior
masonry chimneys and the exterior masonry chimneys. The standard deviations
associated with the mean creosote accumulations for these two chimney types
indicate a potential overlap of the data sets, so caution should be used in
considering the relative ranking of the mean creosote accumulation for the interior
and exterior masonry chimneys. The data set for the exterior masonry chimneys is
relatively small (two values), so additional caution should be used when comparing
this mean creosote accumulation with other mean creosote accumulations.
Conventional Stoves. As in the case of all stove technologies evaluated, the mean
creosote accumulations by chimney type in the conventional stove classification are
ranked as would be anticipated. The prefabricated metal chimneys had the lowest
mean creosote accumulation (0.62 kg/1000 HDD). The interior masonry chimneys had
the middle mean creosote accumulation (0.86 kg/1000 HDD). The exterior masonry
chimneys had the highest mean creosote accumulation (1.63 kg/1000 HDD).
4-75
-------
There are seven or more data points for each chimney type, so a relatively high
level of confidence can be placed in the mean creosote accumulation by chimney type
for the conventional stoves. The difference between the mean creosote
accumulations for the prefabricated metal and interior masonry chimneys is
relatively small (0.24 kg/1000 HDD), and the standard deviations associated with
the respective mean creosote accumulations indicate a potential data set overlap,
so caution should be used in considering the relative ranking of these two chimney
types.
The mean creosote accumulation for the exterior masonry chimneys appears to be
significantly higher than the mean creosote accumulations for the other two chimney
types; however, it is significantly influenced by one data point with a value of
5.78 kg/1000 HDD. With this data point removed from the data set, the resulting
mean creosote accumulation (based on 12 data points) is 1.28 kg/1000 HDD, which is
approximately a factor of two higher than the mean creosote accumulation for the
interior masonry chimneys.
Chimney System Effects on Particulate Emissions
It has been demonstrated that chimney type appears to have a significant effect on
creosote accumulation. It can be further hypothesized that creosote accumulation
may be related to particulate emissions in that relatively higher creosote
accumulation generally indicates higher particulate concentration in the flue gas;
therefore, higher gram-per-hour emission rates would be anticipated to be observed
in conjunction with higher creosote accumulation. This hypothesis is complicated
by the fact that several variables combine to determine a particulate emission
rate, including burn rate, fueling factors, stove design characteristics, etc.
Also, several factors can influence observed creosote accumulation, including
creosote collection methods, chimney type, and creosote removal by pyrolysis during
high burn periods. In summary, if all variables that affect particulate emissions
and creosote accumulation are held constant or their effects minimized, it would be
anticipated that creosote accumulation would be approximately related to
particulate emissions; therefore, the mean particulate emission rates would be
expected to exhibit the same ranking by chimney type as the mean creosote
accumulations.
The number of homes used in the chimney system/emission rate and chimney
system/burn rate analyses is smaller than the number of homes used in the chimney
type/creosote accumulation analysis. Consequently, some chimney types are poorly
represented in some stove technology classifications. In cases where one or two
4-76
-------
homes form the data set for a given chimney type under a technology classification,
the emission rate is significantly influenced by the design of the individual stove
or add-on/retrofit models, so chimney system/emission rate comparisons may be
inappropriate.
Catalytic Stoves. For the catalytic stoves, the interior masonry chimneys had the
lowest mean emission rate (13.6 g/hr). The prefabricated metal chimneys had the
highest mean emission rate (18.2 g/hr). The exterior masonry chimneys had the mid-
range mean emission rate (17.8 g/hr).
The prefabricated metal chimney mean emission rate is from three homes (N01, Stove
A; V31, Stove P; and N32, Stove P). Caution should be used in considering this
mean because it may be a reflection on the emission rate characteristics of the
individual stove models rather than emission rate characteristics for catalytic
stove/prefabricated metal chimney systems.
The mean emission rates for the interior and exterior masonry chimneys are from
twenty or more homes each, so a higher high level of confidence can be placed in
these values. Although the data sets from the masonry chimneys have a relatively
large number of data points, the standard deviations associated with the means
indicate a potential emission rate overlap, so caution should be used when
comparing the mean emission rates for the masonry chimneys.
Add-on/Retrofits. For the add-on/retrofits, the prefabricated metal chimneys had
the lowest "mean" emission rate (7.3 g/hr). This emission rate consists of one
data point from Add-on J, so it should not be considered a representative emission
rate for prefabricated metal chimney/add-on systems. Of the masonry chimneys, the
exterior masonry chimneys had the lowest mean emission rate (16.3 g/hr) and the
interior masonry chimneys had the highest mean emission rate (25.9 g/hr).
As previously mentioned, the add-on retrofit classification had a predominance of
exterior masonry chimneys (63% of the homes). The interior masonry chimney data
set was generated in two homes, both of which used Retrofit F. Therefore, caution
should be used in comparing the mean emission rates for the interior and exterior
masonry chimney systems. The mean emission rate for the interior masonry chimneys
indicates emissions performance of one retrofit, while the mean emission rate for
the exterior masonry chimneys represents a combination of emission rates from
several add-on/retrofit devices.
4-77
-------
Low-emission Stoves. For the low-emission stoves, the interior masonry chimneys
had the lowest mean emission rate (8.8 g/hr). The exterior masonry chimney had the
middle mean emission rate (13.6 g/hr). The prefabricated metal chimneys had the
highest mean emission rate (16.4 g/hr).
The mean emission rate for the prefabricated metal chimneys is significantly
influenced by a single emission rate of 47.6 g/hr (Home V18, Stove K). With this
data point eliminated from the data set, the resulting mean emission rate is 12.9
g/hr (standard deviation 8.1), which would be the middle mean emission rate by
chimney type for the low-emission stoves.
If the 12.9 g/hr mean emission rate is considered representative of the
prefabricated metal chimney systems, the mean emission rates for the three chimney
types may be statistically similar when the standard deviations are taken into
account. The range of the means is relatively narrow (8.8 to 13.6 g/hr), so
caution should be used in considering the relative ranking of mean emission rate by
chimney type for the low-emission stoves. No apparent chimney type/emission rate
correlation is identifiable for the low-emission stove data set.
Conventional Stoves. For the conventional stoves, only masonry chimneys were
evaluated. The interior masonry chimneys had the lowest mean emission rate (10.6
g/hr), and the exterior masonry chimneys had the highest mean emission rate (25.3
g/hr).
The data set for the interior masonry chimneys consists of data from two homes, and
is significantly influenced by the data set from Home V06 (four of five data
points), which had an overall mean emission rate (9.4 g/hr) which was considerably
lower than the overall mean emission rate for all conventional stoves (20.1 g/hr).
Therefore, caution should be used when comparing the mean emission rates by chimney
type for the conventional stoves because the interior masonry mean is significantly
biased by one home where relatively low emission rates were measured.
Chimney System Effects on Burn Rate
It is presumed that the chimney systems with lower heat loss would maintain higher
flue gas temperatures. Additionally, smaller-diameter chimneys would be expected
to create higher draft conditions due to higher gas velocities and lower heat
transfer away from flue gases.
4-78
-------
Prefabricated metal chimney systems are generally located inside the home, and
would be expected to have the lowest heat loss of the three general chimney types.
The interior masonry chimneys would be expected to have relatively higher heat loss
than the prefabricated metal chimneys; however, most heat transfer would be into
the heated area of the home. The majority of heat loss from the exterior masonry
chimneys would be outside the home.
Catalytic Stoves. For catalytic stoves, the prefabricated metal chimneys had the
lowest mean burn rate (0.85 kg/hr), the interior masonry chimneys had the middle
mean burn rate (1.02 kg/hr), and the exterior masonry chimneys had the highest mean
burn rate (1.30 kg/hr).
The mean burn rates for the three chimney types appear to be significantly
different. The mean burn rates for the prefabricated metal and exterior masonry
chimneys do not overlap when the standard deviations for these data sets are
considered. The mean burn rate for the interior masonry chimneys is approximately
mid-way between the mean burn rates for the prefabricated metal and masonry
chimneys.
Add-on/Retrofits. The lowest overall mean burn rate for the add-on/retrofit
classification was for the interior masonry chimneys (1.15 kg/hr). The middle
"mean" burn rate was for the prefabricated metal chimneys (1.31 kg/hr); however, as
in the case of the chimney system/emission rate analysis, this "mean" burn rate
consists of a single data point. The highest mean burn rate for the add-on/
retrofits was for the exterior masonry chimneys (1.69 kg/hr).
As in the case of the chimney system/emission rate analysis, caution should be used
in comparing mean burn rates by chimney system type for the add-on/retrofits. The
data set for the interior masonry chimneys was generated in two homes using
Retrofit F, so the mean burn rate for this chimney type should not be considered
representative of a mean burn rate for interior masonry chimneys with add-on/
retrofits.
Low-emission Stoves. The interior masonry chimneys had the lowest mean burn rate
for the low-emission stoves (0.98 kg/hr). The exterior masonry chimneys had the
middle mean burn rate (1.01 kg/hr). The prefabricated metal chimneys had the
highest mean burn rate (1.06 kg/hr).
4-79
-------
As previously discussed, the low-emission stoves evaluated in this study appear to
have a characteristically narrow range of burn rates (0.67 to 1.38 kg/hr for
individual sampling periods). This characteristic is reflected in the mean burn
rates by chimney types.
The range of the three mean burn rates by chimney types for the low-emission stoves
is relatively narrow. When the standard deviations associated with each mean are
taken into account, it appears that the differences in the mean burn rates by
chimney type are statistically insignificant. This observation indicates that
chimney type (and associated heat transfer characteristics of the chimney type)
exerts a relatively small influence on burn rate for low-emission stoves.
Conventional Stoves. As in the case of the chimney system/emission rate analysis,
only masonry chimneys were evaluated with the conventional stoves. The interior
masonry chimneys had the lowest mean burn rate (1.69 kg/hr), and the exterior
masonry chimneys had the highest mean burn rate (1.76 kg/hr).
The difference in the two means for the masonry chimney/conventional stove systems
(0.07 kg/hr) is statistically insignificant when the standard deviation associated
with each mean is taken into account. This indicates that for the conventional
stoves evaluated in this study, chimney system had a relatively small influence on
burn rate.
FIREBOX SIZE EFFECTS
Based on previous research (_5) and preliminary review of data from this study, the
effects of firebox size on particulate emissions were investigated. Each stove
technology category was reviewed individually as well as all stoves combined.
Catalytic Stoves.
Firebox volumes were compared with mean particulate emission rates for all
catalytic stoves, both those provided to the study (A, B, C, D) and existing units
(P). The resulting plot (Figure 4-11A) and r2 value (0.651) indicates a relatively
good correlation, considering all the factors which can affect emission rates.
The relationship between catalytic stove firebox size and emissions is not apparent
from laboratory certification and screening tests for these stoves. However,
4-80
-------
Figure 4-11A
Particulate Emissions (g/hr) vs. Firebox Size (liters): Catalytic Stoves
35 -
30 -
HEftN
PflRTICULflTE
EMISSIONS
(G/HR)
I
CO
20 -
10 -
5 -
r*
= 0
€51
r
20
I
40
SO 80
FIREBDH SIZE (LITERS)
100
120
NOTE: Each data point riprtJtntJ the Htsn uslut for a stout Hodsl Sin^lt ualues art shown without
standard deuiations. flultiplt data points art shown for stout typt P dut to uaryin$ firtbow sizts.
-------
considering the relatively low values for percent of time combustors were "lit
off," and a positive correlation between conventional stove firebox size and
emissions (see Figure 4-11D2), it appears that the catalytic stoves are operating
as conventional stoves (without catalytic action) for a significant amount of the
time they are burned. This is further supported by the slopes of catalytic and
conventional stove plots; the slope of the catalytic stove curve is shallower,
showing lower emissions than conventional stoves for the same firebox size.
Add-on/Retrofits.
Figure 4-11B shows mixed results (r2 = 0.329) for add-on/retrofit devices. The
poorer correlation is probably due to a narrower range of stove sizes (the smallest
firebox in this group is larger than 60 liters) and the relatively good performance
of Stove E (a retrofit). Results were computed without one Stove J (add-on) sample
which appeared to be an outlier. Although most of the add-on/retrofit devices were
installed on stoves with a narrow range of firebox volumes (74 to 84 liters),
emissions in this size range are similar to those observed in the conventional
stove group.
Low-Emissions Stoves.
Although the low-emission stoves show an r2 value of 0.354, no real correlation is
apparent (Figure 4-11C). This is due to the very narrow range of firebox sizes (37
to 49 liters), which makes detecting a trend difficult. However, emissions from
the low-emission stoves are similar to or lower than emissions from comparably-
sized conventional stoves. Results were computed without on Stove K sample, which
appeared to be an outlier.
Conventional Stoves.
The "firebox size hypothesis" as a major factor in woodstove emission performance
was originally based on the laboratory testing of conventional stove technology
referenced above. The conventional stove field testing data presented here clearly
confirms the hypothesis. Figure 4-11D1 shows emissions from conventional stoves
compared to the actual firebox volumes of the stoves. One stove appears to be an
outlier, with emissions about half that indicated by similarly-sized stoves.
However, this stove was always operated with a very deep ash bed which effectively
reduced the firebox volume by half. When this stove is replotted by its effective
firebox volume (Figure 4-11D2), the r2 value increases from 0.349 to 0.722. Based
on the unusual ash bed in this stove, the effective firebox volume is considered
appropriate and is used in the following evaluation of all stoves.
4-82
-------
Figure 4-liB
Participate Emissions (g/hr) vs. Firebox Size (liters): Add-On/Retrofits
35--
30 -
HERN
PRRTICULflTE
EHISSIQNS
(G/HR)
i
co
15 -
10 -
20
I
HO
60 80
FIREBDK SI2E (LITERS)
I
100
120
NOTE: Each data point rtprtitntj tht Hcan uslut for a stout Hodt l. Siri^lc ualuti art shown without standard
dtuiitions. Multiplt data points act shown for add-on/rttrofit typss G, I, and J dut to uaryin^ flrtbOK sizts,
-------
Figure 4-11C
Particulate Emissions (g/hr) vs. Firebox Size (liters): Low Emission Stoves
30 -
HERN
PflRTICULflTE
EMISSIONS
(G/HR)
I
Oo
10 -
r£ =0
35H
I
20
I
40
I
60
1
80
100
120
FIREBDK SIZE (LITERS)
NDTE: Each data point rtprtstnts tht Htan ualut for a stout Hodti.
Tht Hem sHission ualut for stout K without hont U18 would bt 11.2
Sin^lt ualuts art shown without standard dtulationi
with a standard dtuiation of 1.8 3/hr
-------
Figure 4-ilDl
Participate Emissions (g/hr) vs. Firebox Size (liters): Conventional Stoves
35 -
30 -
HERN
PflRTICULflTE
EMISSIONS
(fl/HR)
I
00
tn
10 -
5 -
r*
= 0
3H9
I
20
r
HO
i
60
80
100
120
FIREBDH SIZE (LITERS)
NOTE: Each data point represents tht Hcsn ualus for a stout Hodtl. Single ualues are shown without
standard dsulations. Multiple data points art shown for stout type D dut to uaryin^ firtbow sizes.
-------
Figure 4-11D2
Participate Emissions (g/hr) vs. Firebox Size (liters): Conventional Stoves
30 -
HEftM
PfiRTICULfiTE
EHISSIONS
(G/HR)
do
cn
10 -
5 -
r2 =o
722
I
20
l
HO
I
60
I
80
100
120
FIREBOK SIZE (LITERS)
HD1E: Each data point represents the Hear
an ualut for a stoue Hode I . Single ualues are shown without
standard deviations. Multiple data points are shown for stoue type D due to uaryin^ firebOH sizes.
-------
AH Stoves.
All study stoves, including new and existing stoves, were evaluated together for
firebox size effects. The evaluation was made without the outlying Stove J sample
and using the effective firebox volume for the Stove 0 installation mentioned
above. A single Stove K value (V18-7) was also deleted as atypical. The resulting
positive-slope curve has an r2 value of 0.475 (Figure 4-11E). Although there is a
reasonable amount of scatter, it is remarkable that the correlation is so strong
despite the effects of stove technology and other factors. The fundamental
combustion conditions found in small fireboxes (limited size of fuel loads, greater
turbulence and mixing, and higher firebox temperatures) apparently are at least as
important as other factors.
ADVANCED TECHNOLOGY STOVE ANALYSIS
The following discussion is focused on the overall performance of the advanced
technology stove models (catalytic stoves, add-on/retrofits, and low-emission
stoves). Particulate emissions are a primary regulatory concern, and factors that
appear to influence emission rates for individual stove models are therefore
emphasized. The analysis concentrates primarily on those sampling periods during
which emissions samples were obtained.
The creosote accumulation discussion is limited to those homes that underwent
emissions sampling (Groups I and III). This approach was adopted in order to
simplify the discussion and to keep the analysis focused on factors affecting
particulate emissions.
From the consumer perspective, there are several factors in addition to emissions
performance that are important in an advanced technology stove, including the range
of heat output, convenience of operation, safety, durability, and fuel
efficiencies. The following discussion addresses these factors and includes
comment from the study participants to give an overall evaluation of each stove
model.
Catalytic Stoves
Stove A. Stove A was evaluated in two homes (N01 and N10). The overall mean
emission rates for the two homes differed by 4.8 g/hr (18.0 g/hr for Home N01, 22.8
g/hr for Home.N10).
4-87
-------
Figure 4-11E
Particulate Emissions (g/hr) vs. Firebox Size (liters): All Stoves
30 -
HERN
PflRTICULflTE
EMISSIONS
(G/HR)
CO
CD
20 -
10 -
r? =0
t75
20
HO
]
60
I
80
100
120
FIREBOH SIZE (LITERS)
NOTE: Each data point rtprtstnts tht Htsn ualut far a stout nodtl Sin^lt ualutj art shown without standard
dtuistions. Multiple data points art shown for stout typts ?, G, I, J, and D dut to uaryin^ firtbon s\xts.
-------
The average fuel loads, fueling frequencies, burn rates, and catalyst operation
times were significantly different in the two homes. Home N01 had an overall mean
fuel load of 4.5 kg, an overall mean fueling frequency of 0.15 #/hr, an overall
mean burn rate of 0.70 kg/hr, and an overall mean catalyst operation time of 26.0%.
Home N10 had an overall mean fuel load of 7.2 kg, an overall mean fueling frequency
of 0.21 l/hr, an overall mean burn rate of 1.42 kg/hr, and an overall mean catalyst
operation time of 86.3%. Home N10 had larger fuel loads, more frequent fueling,
and higher burn rates. Presumably, the higher burn rate in N10 also contributed to
the higher catalyst operation time.
Although the fuel consumption characteristics and catalyst operation times were
significantly different in the two homes, the mean emission rates were not
significantly different when the standard deviations associated with each mean
emission rate are taken into account. Home N01 had a narrower range of emission
rates for individual sampling periods (13.0 to 21.9 g/hr), but its range is
contained within the range of emission rates for Home N10 (9.7 to 39.7 g/hr).
There were two sampling periods for Stove A where relatively low emission rates
(less than 14.0 g/hr) were recorded. Sampling period N01-3 had an emission rate of
13.0 g/hr, and sampling period N10-1 had an emission rate of 9.7 g/hr. When the
complete data set from the individual homes is analyzed, there is no common factor
between the two homes which would clearly explain the lower observed emission rates
during these sampling periods. It appears that the factors which potentially
affect emission rates combined during these sampling periods to produce the lower
emission rates.
In Home N10 there was one sampling period where a relatively high (higher than 25.0
g/hr) emission rate was measured (N10-7, 39.7 g/hr). The average fuel load for
this sampling period was 11.1 kg, which was 3.0 kg higher than the second highest
average fuel load in this home (8.1 kg for N10-6). Along with the highest average
fuel load, the lowest fuel loading frequency was observed during this sampling
period (0.14 #/hr). All other stove operation characteristics do not appear to be
outside of the normal range of values observed in this home.
The data set from Stove A does not give a conclusive indication of factors that
affect observed emission rates. On the contrary, it appears that the emission rate
is relatively unaffected by burn rate, catalyst operation time, and fueling
4-89
-------
practices. It appears that Stove A is capable of both relatively high and
relatively low emission rates.
The chimney systems were significantly different in the two homes, which was
reflected by the observed creosote accumulations. Home N01 had a prefabricated
metal chimney, and a mean creosote accumulation (two heating seasons) of 0.61
kg/1000 HDD. Home N10 had an exterior masonry chimney and a mean creosote
accumulation (two heating seasons) of 1.15 kg/1000 HDD.
Participants in each of the homes where Stove A was evaluated reported problems
with creosote condensate in the chimney system. Home N01 reported that icicles
would form at the chimney exit on colder days. Home N10 reported that creosote
condensate had leached through the masonry blocks of the chimney. These
observations are not necessarily due to operation characteristics of Stove A
because each home used relatively wet fuel. The fuel moisture content for
individual sampling periods in Home N01 ranged from 29.0% to 43.0% (mean 36.3%).
The fuel moisture content for individual sampling periods in Home N10 ranged from
26.0% to 41.4% (mean 36.2%). It is probable that if drier fuel was used in these
homes, the flue condensation problems would decrease.
Stove B. Stove B was evaluated in four homes (V05, Vll, N09, and N18). The
measured emission rates ranged from 6.1 to 41.3 g/hr for individual sampling
periods, and the overall mean emission rates varied significantly in the homes
where Stove B was evaluated. Home V05 had an overall mean emission rate of 20.2
g/hr, Home Vll had an overall mean emission rate of 6.5 g/hr, Home N09 had an
overall mean emission rate of 20.9 g/hr, and Home N18 had an overall mean emission
rate of 30.7 g/hr.
The mean emission rates for Stove B in Homes V05, N09, and N18 can be considered
statistically similar when the standard deviations associated with each emission
rate are taken into account. The mean emission rate for Home Vll is significantly
lower than the mean emission rates from the other three homes. The mean burn rate
(1.12 kg/hr) and burn rate range (1.02 to 1.19 kg/hr) for Home Vll were in the
normal range of burn rates observed for Stove B. However, the fueling pattern in
Home Vll was significantly different than in the other Stove B homes. Home Vll had
the highest overall mean fuel load (12.7 kg), the lowest overall mean fuel loading
frequency (0.09 #/hr), and the lowest overall mean catalyst operation time (54.1%).
A review of the data files from this home indicates that the stove was usually
fueled with a relatively large fuel load and allowed to burn out prior to
4-90
-------
refueling. The other three homes where Stove B was evaluated tended to frequently
fuel the stove and maintain fairly constant burn rates. The fueling pattern in
Home Vll probably resulted in lower measured emission rates because of the
relatively long "charcoal phase" at the end of each burn cycle when emissions are
generally low. The presence of the "charcoal phase" (and associated lower flue gas
temperatures) in Home Vll probably also accounts for the low catalyst operation
time. The burn cycles for the other three homes did not generally have "charcoal
phase" periods.
Home V05 did have one sampling period (V05-4) where a relatively low emission rate
was recorded (9.0 g/hr). This sampling period was accompanied by the smallest
average fuel load (4.4 kg), highest fuel loading frequency (0.19 #/hr), and lowest
fuel moisture (18.2%) recorded in this home. Although these factors are probably
significant, it is difficult to conclusively identify which factors contributed to
the lower emission rate measured during this period.
The data set from Stove B contains five sampling periods with relatively high
emission rates. These sampling periods include V05-5 (31.4 g/hr), N09-7 (29.6
g/hr), N18-5 (41.3 g/hr), N18-6 (31.6 g/hr), and N18-7 (29.2 g/hr). Review of the
data sets for the remaining five sampling periods does not indicate any significant
factors that would appear to contribute to the relatively high measured emission
rates measured. It appears that the Stove B emission control system (bypass
damper, secondary air, and catalyst) may be relatively sensitive. Factors which
could contribute to higher particulate emissions but may not be observable in the
data sets include an unsealed bypass damper, an unsealed catalyst, or an
ineffective catalyst.
Three homes using Stove B had exterior masonry chimneys (V05, N09, and N18). Home
Vll had an interior masonry chimney and had the lowest mean creosote accumulation
(0.44 kg/1000 HDD over two heating seasons). Home V05 had a creosote accumulation
of 1.03 kg/1000 HDD (one heating season), Home N09 had 1.13 kg/1000 HDD (two
heating seasons), and Home N18 had 0.54 kg/1000 HDD (one heating season).
Each of the four homes where Stove B was evaluated reported varying degrees of
creosote condensation problems. In Home V05 creosote condensate leached into the
masonry chimney blocks along the entire length of the chimney system. Ice formed
in the lower section of the chimney during colder days. The fuel used in V05 did
not appear to have a notably high moisture content (mean of 23.7%, range of 18.2%
to 29.5%). The exterior masonry chimney was insulated and was extended from 6.7
4-9.1
-------
meters (22 feet) to 7.3 meters (24 feet) between heating seasons; however, the
creosote condensate leaching and excess water/ice problem persisted during the
second heating season.
Home Vll also reported creosote condensate leaching through the masonry chimney
blocks and excess water collecting at the base of the chimney. The chimney was
located indoors, so the creosote condensate caused odor problems. The chimney was
rebuilt between heating seasons; however, the water problem persisted. As in the
case of V05, the fuel used in Vll did not appear to appear to have a notably high
moisture content (mean of 26.3%, range of 24.0% to 28.0%).
Home N09 also reported creosote condensate leaching into the masonry chimney blocks
and excess water in the chimney which caused ice formation near the clean-out door
at the base of the chimney. The homeowner placed a light bulb in the clean-out
door to prevent ice from forming inside the chimney. The fuel moisture content was
significantly reduced during the second heating season, and this appeared to reduce
the amount of water collecting in the base of the chimney. During the first
heating season the fuel moisture content was 41.0% (one sampling period). During
the second period the mean fuel moisture content was 16.5%, with a range of 15.8%
to 17.1%.
Home N18 reported some creosote condensate leaching into the masonry chimney
blocks; however, no water problem was reported. The moisture content of the fuel
used in this home was fairly low (mean of 15.5%, range of 11.0% to 17.8%).
Stove C. Stove C was evaluated in three homes (V07, V16, and N03), with mean
emission rates of 9.4 g/hr (V07), 16.2 g/hr (V16), and 17.1 g/hr (N03).
The mean emission rate for Home V07 is significantly influenced by one sampling
period (V07-7), where the measured emission rate was 1.7 g/hr. If this sampling
period is eliminated from the data set, the resulting mean emission rate for this
home is 11.9 g/hr. A review of the data set for this home does not indicate any
significant factors that would have contributed to the low emission rate measured
during sampling period V07-7; this sample may be invalid.
Along with the lowest overall mean emission rate for Stove C, V07 also had the
highest overall mean burn rate (1.60 kg/hr), highest overall mean catalyst
operation time (71.9%), largest overall mean fuel load (9.1 kg), and lowest overall
mean fueling frequency (0.16 #/hr). The above factors all appear to have combined
4-92
-------
to achieve the lowest overall mean emission rate for this stove. The homeowner
appeared to use a relatively large fuel load and operated the stove at a relatively
high burn rate, which probably contributed to high catalyst operation times.
In Home V16, the catalyst operation time appeared to be decreased during the second
heating season. The bypass linkage rod was coated with brown material in the
second heating season, which, according to the stove operation manual, may indicate
an inactive catalyst. The mean catalyst operation time decreased from 80.2% (one
sampling period) during the 1985-86 heating season to 59.3% (four sampling periods)
during the 1986-1987 heating season. There was also an increase in the moisture
content of the fuel used in this home from 18.5% (one sampling period) during the
1985-86 heating season to 28.3% (four sampling periods) during the 1986-1987
heating season. The mean particulate emission rate increased along with the
decreased catalyst operation time and increased fuel from 8.2 g/hr (one sampling
period) during the 1985-86 heating season to 18.2 g/hr (four sampling periods)
during the 1986-87 heating season.
As in the case of Home V16, Home N03 also had one sampling period (N03-4) where the
measured emission rate (8.1 g/hr) was significantly lower than the emission rates
for the other sampling periods (19.0 g/hr for N03-5, 24.3 g/hr for N03-6). A
review of the data sets indicates that the stove was used less in sampling period
N03-4 (40.6% operation time) than in the other sampling periods (76.5% in N03-5,
66.4% in N03-6). The relatively low measured emission rate in sampling period N03-
4 may be an artifact of the stove being allowed to burn out during this sampling
period. During sampling periods with relatively low "stove operation time," or
with long periods between refueling, a proportionately larger percentage of the
S '
sampling period would include the "charcoal phase" of the burn cycle, which would
be expected to reduce the measured emissions.
All of the homes where Stove C was evaluated had masonry chimney systems; Homes V07
and V16 had exterior masonry chimneys, and Home N03 had an interior masonry
chimney. The mean creosote accumulations (two heating seasons in each home) for
the three homes were 0.60 kg/1000 HDD for Home V07, 0.79 kg/1000 HDD for Home V16,
and 0.29 kg/1000 HDD for Home N03, fitting the pattern observed before.
The homeowners expressed varying degrees of satisfaction with Stove C. The
participants in V07 reported that the heat output of the stove was insufficient.
The participants in V16 were very pleased with the performance of Stove C. The
participant in N03 also was very pleased with the performance of Stove C; however,
4-93
-------
this home also reported that creosote condensate occasionally dripped from the
clean-out door at the base of the chimney.
Stove D. Stove D was evaluated in four homes (V08, V13, N02, and Nil). The mean
emission rates in the four homes were 14.4 g/hr (V08), 12.9 g/hr (V13), 8.4 g/hr
(N02), and 10.2 g/hr (Nil). As discussed in Section 3 of this report and
summarized in Tables 3-14 and 3-16, the cordierite-based catalysts used in the
1985-86 heating season were changed to mullite-based catalysts for the 1986-87
heating season, which decreased the overall mean emission rate in all of the four
homes during the second heating season.
Relatively low emission rates were observed in one or more sampling periods in each
of the homes where Stove D was evaluated. For the overall Stove D data set, 15 of
19 individual sampling periods had measured emission rates that were less than 15.0
g/hr. Twelve of the 15 sampling periods with relatively low emission rates
occurred during the second heating season with the mullite-based catalysts in
place, while all four sampling periods with emission rates above 15.0 g/hr occurred
during the first heating season with the cordierite-based catalysts in place.
Home N02 was the only home where Stove D was evaluated that had emission rates less
than 10.0 g/hr during the first heating season (9.9 g/hr, N02-1 and 7.0 g/hr, N02-
3). During these periods the two highest catalyst operational times for individual
Stove D sampling periods were recorded (83.1% for N02-1, 74.1% for N02-3).
Based on the above data, the mullite-based catalysts appear to generally produce
lower emission rates in Stove D than the cordierite-based catalysts; however, if
catalyst activity is maintained, the cordierite-based catalysts also appear to be
capable of relatively low emission rates. The lower emission rates observed during
the second heating season may simply reflect a larger fraction of time that
combustor substrates were in good condition.
The mean burn rates in each of the four homes where Stove D was evaluated were
similar and overlap. The mean burn rate range was from 0.89 kg/hr (Nil) to 1.11
kg/hr (V13). This parameter may be significantly influenced by the Stove D design,
which had the smallest firebox (38 liters, 1.4 cubic feet) of all catalytic stove
models.
The highest Stove D emission rate measured during an individual sampling period was
20.4 g/hr for sampling period V08-2. This is an indication that the emission
4-94
-------
control system of Stove D may be relatively insensitive to factors which can cause
higher emissions in other stoves. It may be that if the catalytic system does not
work, emissions are still relatively low due to the small firebox size.
Homes V08, V13, and N02 all appeared to exhibit similar fueling habits. The
overall mean fuel loads for these homes were 5.0 kg, 4.0 kg, and 5.4 kg,
respectively. The overall mean fueling frequencies for these homes were 0.20,
0.27, and 0.19 #/hr, respectively. Home Nil had the highest overall mean fueling
frequency, 0.33 #/hr; however, this fueling frequency was not significantly higher
than the fueling frequency for Home V13. Home Nil did have a significantly smaller
overall mean fuel load than the other three homes (2.7 kg). The smaller fuel load
and higher fueling frequency in Home Nil did not appear to significantly affect the
observed emission rates.
The design of Stove D appears to inherently minimize the effects of operator
factors. As demonstrated by Home Nil, fueling habits do not appear to affect the
emission rate. The emission rates for this stove model as evaluated in this study
appeared to be primarily affected by catalyst type rather than operator factors.
Three of the four homes where Stove D was evaluated had interior masonry chimney
systems. Home N02 had an exterior masonry chimney system. The mean creosote
accumulations (two heating seasons in each home) for the individual homes were 0.64
kg/1000 HDD (V08), 0.35 kg/1000 HDD (V13), 0.73 kg/1000 HDD (N02), and 0.50 kg/1000
HDD (Nil).
The participants in the homes reported varying degrees of satisfaction with Stove
D. The participants in Home V13 were very pleased with the operation of the stove.
The participants in Homes V08 and N02 did not comment on the stove's performance;
however, Home V08 experienced three chimney fires during the study. The
participants in Home Nil reported that the bypass became difficult to operate when
the stove was very warm. The participants in Home Nil also expressed
dissatisfaction with the type heat output of the stove; they would have preferred a
convection-type heater (which they had prior to the study) rather than a radiant-
type heater.
Stove Code P. Stove Code P is comprised of six catalytic stoves that had been in
use for at least one heating season prior to the start of the study. Four homes
had one or more valid emission sampling periods.
4-9.5
-------
One sample was completed in Home V31. This catalytic stove had the second smallest
firebox size (after Stove D) of all catalytic stoves (40 liters, 1.4 cubic feet).
The emission rate was 17.7 g/hr, which was 1.1 g/hr higher than the overall mean
emission rate for all catalytic stoves of 16.6 g/hr-
Two sampling periods were completed in Home V32. This stove model also had a
relatively small firebox size (52 liters, 1.8 cubic feet). It is similar to Stove
C but is smaller and has no secondary air. The two emission rates measured for
this stove were 13.9 g/hr (V32-1) and 11.8 g/hr (V32-5). The study participant in
this home replaced the catalyst between the two heating seasons; however, the data
set is too small to formulate any meaningful conclusions regarding the results of
the catalyst change. The two emission rates measured for this stove model were
lower than the overall mean emission rate for all catalytic stoves of 16.6 g/hr.
One sampling period was completed in Home N32. This catalytic stove model is
essentially the same stove as Stove A; however, the catalyst in this stove is 8 cm
(3 inches) thick, while the catalysts in the other Stove A models were 5 cm (2
inches) thick. The single emission rate measured in Home N32 (19.6 g/hr) was
essentially identical to the overall mean for Stove A (20.4 g/hr), and 3.0 g/hr
higher than the overall mean for all catalytic stoves (16.6 g/hr). The difference
between the emission rate in Home N32 and the overall mean for Stove A is
statistically insignificant.
Two sampling periods were completed in Home N33. This stove model had the largest
firebox size of the existing catalytic stoves (119 liters, 4.2 cubic feet). The
two emission rates measured for this stove were 22.3 and 34.6 g/hr, both of which
are higher than the overall mean for all catalytic stoves of 16.6 g/hr. This stove
was operated at relatively high burn rates; 1.83 kg/hr with the 22.3 g/hr emission
rate, and 2.26 kg/hr (the highest burn rate measured for all catalytic stoves) with
the 34.6 g/hr emission rate. The moisture content of the fuel burned in this home
was relatively high (mean of 31.2%). At the conclusion of the study (March 1987)
the catalyst was replaced, and the homeowner reported improved stove performance,
which may indicate that an improperly operating catalyst could have been in place
during the emission sampling periods.
Add-on/Retrofits
Retrofit E. Retrofit E was evaluated in one home (V01) during the 1986-87 heating
season. The overall mean emission rate for this device was 7.8 g/hr, with a range
of 6.3 to 10.1 g/hr for three sampling periods.
4-96
-------
The same manufacturer produced both the retrofit device and the conventional stove
on which the retrofit was installed, although this retrofit can be installed on
other makes of conventional stoves with similar features. This conventional stove
had a firebox size of 62 liters (2.2 cubic feet).
The emission rates recorded for Retrofit E were all relatively low. The burn rate,
catalyst operation time, and fueling characteristics did not vary significantly for
the three sampling periods. For individual sampling periods the burn rate range
was 1.17 to 1.37 kg/hr, the catalyst operation time range was 63.9% to 74.7%, the
average fuel load range was 5.3 to 6.4 kg, and the average fuel loading frequency
range was 0.21 to 0.22 #/hr. The catalyst in Retrofit E is held in a cast iron
housing, which may have helped keep combustor temperature high during refueling
periods.
Retrofit E, as used in Home V01, had emission rates similar to the better-
performing catalytic and low-emission stoves. Caution should be used in forming
conclusions regarding Retrofit E due to the relatively small data set from a single
home. Within this home the fueling and burn rate characteristics were not observed
to vary significantly, so the relatively low emission rates observed for this
retrofit model may be indicative of emissions performance under a limited range of
operating conditions.
Home V01 had an exterior masonry chimney. The mean creosote accumulation (two
heating seasons) was 0.49 kg/1000 HDD. This mean creosote accumulation is at the
lower end of the range for catalytic technology with exterior masonry chimneys.
The study participants in Home V01 did not comment on the performance of Retrofit
E.
Retrofit F. Retrofit F was evaluated in two homes (V03 and V12) during the 1985-
86 heating season. The overall mean emission rates for these two homes were 25.2
g/hr (V03) and 26.6 g/hr (V12).
Retrofit F is designed for installation on a specific conventional stove model;
however, the retrofit is no longer being manufactured. The device was installed on
a conventional stove with a firebox volume of 74 liters (2.6 cubic feet).
4-97
-------
Two sampling periods were completed in each home where Retrofit F was evaluated, so
the overall means for each home consist of only two values. The emission rates for
the individual homes varied significantly during the two sampling periods. The
emission rates for individual sampling periods in Home V03 were 18.6 g/hr (V03-1)
and 31.8 g/hr (V03-3). The emission rates for individual sampling periods in Home
V12 were 16.5 g/hr (V12-2) and 36.7 g/hr (V12-3).
For sampling period V03-1, the emission rate was 18.6 g/hr while the catalyst
operation time was 25.5%. For sampling period V03-3, the emission rate was 31.8
g/hr while the catalyst operation time was 17.7%. For sampling period V12-3, the
emission rate was 36.7 g/hr while the catalyst operation time was 9.8%. Due to a
thermocouple failure there was no catalyst operation time calculated for sampling
period V12-2. Apparently Retrofit F is capable of achieving relatively high
emission rates when catalyst activity is not maintained.
The fueling and burn rate characteristics in each home did not conclusively
indicate any relationship to the measured emission rates. This may be due to the
relatively small data set for Retrofit F.
Both of the homes where Retrofit F was evaluated had interior masonry chimneys.
The creosote accumulations (one heating season) for these two homes were 0.66
kg/1000 HDD (V03) and 0.73 kg/1000 HDD.
The study participants in Home V03 reported smoke intrusion into the home on at
least four occasions while the stove was left unattended. The smoke had exited the
stove via the secondary air inlet on the retrofit housing. The study participants
in Home V12 did not comment on the operation of Retrofit F.
Add-on 6. Add-on G was evaluated in two homes (V02 and N04). A total of three
emission samples were obtained for this add-on. The mean emission rate for Home
V02 (two sampling periods) was 16.3 g/hr (emission rates during individual sampling
periods of 15.5 and 17.1 g/hr). The emission rate for Home V12 (one sampling
period) was 18.7 g/hr.
The firebox sizes of the conventional stoves on which Add-on G was installed were
77 liters (2.7 cubic feet) in Home V02 and 84 liters (3.0 cubic feet) in Home N04.
The difference in the three emission rates for Add-on G was relatively small (all
within 1.6 g/hr), which may be either an indication that Add-on G is capable of
4-98
-------
consistent emissions performance when installed on a range of conventional stove
sizes, or an artifact of a small data set.
A review of the Add-on G data set indicates that the burn rates and catalyst
operation times did not vary significantly between Homes V02 and N04. The overall
average fuel load (two sampling periods) was larger in Home V02 (11.3 kg) than the
average fuel load (one sampling period) in Home N04 (8.2 kg). As would be
anticipated based on the fuel load size relationship between homes, the mean
overall fuel loading frequency (two sampling periods) was smaller in Home V02 (0.14
#/hr) than the fuel loading frequency (one sampling period) in Home N04 (0.21
#/hr).
Each of the two homes where Add-on G was evaluated had exterior masonry chimney
systems. Home V02 had a mean creosote accumulation (two heating seasons) of 0.71
kg/1000 HDD. Home N04 did not use Add-on G for an entire heating season, so a
creosote accumulation representative of Add-on G for this home is not available.
Study participants in each of the two homes where Add-on G was evaluated reported
that they were very pleased with the operation of the add-on.
Add-on H. Only one emission sample (in Home V10) was obtained for Add-on H. The
emission rate for this sample was 16.2 g/hr.
It is difficult to speculate on the overall performance of Add-on H based on a
single emission sample; 16.2 g/hr may be low, high, or average for Add-on H.
Home V10 had an exterior masonry chimney. The creosote accumulation (one heating
season) was 0.41 kg/1000 HDD, which is at the lower end of the range for add-on/
retrofits installed in conjunction with exterior masonry chimneys.
The study participants in Home V10 reported catalyst ash plugging and smoke
spillage from the add-on unit housing.
Add-on I. Add-on I was evaluated in two homes (N06 and N14). The mean emission
rate for Home N06 was 22.6 g/hr (three sampling periods). The emission rate (one
sampling period) for Home N14 was 25.7 g/hr.
4-99
-------
The firebox sizes of the conventional stoves on which Add-on I was installed were
78 liters (2.8 cubic feet) for Home N06, and 119 liters (4.2 cubic feet) for Home
N14.
The emission rates for individual sampling periods in Home N06 varied significantly
(range of 13.6 to 37.3 g/hr). A review of the data set from sampling period N06-3
(when the 37.3 g/hr emission rate was recorded) does not indicate any significant
parameters that would appear to contribute to the relatively high emission rate
recorded during this sampling period.
Relatively high burn rates (range of 2.08 to 2.35 kg/hr) were measured in each of
the four sampling periods completed on Add-on I. These burn rates were
significantly higher than the next highest add-on/retrofit burn rate which had an
associated emission sample (1.70 kg/hr, Home N04, Add-on G).
Each of the two homes where Add-on I was evaluated had exterior masonry chimney
systems. The creosote accumulations (one heating season in each home) were 0.46
kg/1000 HDD (N06) and 1.64 kg/1000 HDD (N14). The significant difference in
measured creosote accumulation (1.18 kg/1000 HDD) should be interpreted with
caution. The relatively high burn rates measured in each of the homes may have
contributed to a lower deposition rate or creosote loss by pyrolysis. Home N06 had
an overall mean flue gas temperature of 257°C (494°F), which was the highest mean
flue gas temperature observed for all add-on/retrofits. Home N14 had an average
(one sampling period) flue gas temperature of 218°C (424°F). Both chimney systems
were similar.
The study participants in Home N06 did not comment on the operation of Add-on I.
The study participants in Home N14 reported observing a soot-plugged catalyst, but
did not comment on the performance of this device.
Add-on J. Add-on J was added to the study for the second heating season, as it had
shown the best emission reduction performance of add-on devices in laboratory
testing. Add-on J was evaluated in three homes (V10, N04, and N12) during the
1986-87 heating season; however, only four sampling periods were completed among
the three homes (two sampling periods in Home V10, one sampling period in Homes N04
and N12). The "mean" emission rates for Add-on J were 14.9 g/hr (two samples of
8.4 and 21.3 g/hr) in Home V10, 14.2 g/hr in Home N04, and 7.3 g/hr in Home N12.
4-100
-------
The firebox sizes of the conventional stoves on which Add-on J was installed were
81 liters (2.9 cubic feet) in Home V10, 84 liters (3.0 cubic feet) in Home N04, and
111 liters (3.9 cubic feet) in Home N12.
The highest emission rate measured for Add-on J (21.3 g/hr, sampling period V10-5)
was accompanied by the lowest catalyst operation time measured for this add-on
(17.6%). The study participant was observed to be operating the add-on in the
partially bypassed position during this sampling period, which would account for
the relatively low catalyst operation time and the relatively high emission rate.
The homeowner fully engaged the catalyst for sampling period V10-6, resulting in a
catalyst operation time of 56.2% and an emission rate of 8.4 g/hr.
Add-on J had the lowest overall mean emission rate of all add-ons (12.8 g/hr). If
sampling period V10-5 is eliminated from the data set because of improper operation
of the device, the overall mean emission rate for Add-on J would be 10.0 g/hr.
This is an indication that the add-on may be capable of reducing emissions from
conventional stoves (based on an overall mean emission rate for conventional stoves
of 20.1 g/hr).
The relative ranking by chimney type of the creosote accumulations in the homes
where Add-on J was evaluated are as would be expected. Homes V10 and N04 had
exterior masonry chimney systems, and the creosote accumulations (one heating
season in each home) were 0.49 and 2.90 kg/1000 HDD, respectively. Home N12 had a
prefabricated metal chimney system and a creosote accumulation (one heating season)
of 0.20 kg/1000 HDD.
Although the creosote accumulations are ranked as would be expected by chimney
type, there is a significant difference (2.41 kg/1000 HDD) in the creosote
accumulations for the two exterior masonry chimneys. Home V10 had an average flue
gas temperature (two sampling periods) of 103°C (217°F), while Home N04 had an
average flue gas temperature of 226°C (438°F). This appears to contradict
conventional wisdom, where higher flue gas temperatures would be expected to result
in relatively less creosote accumulation. Home N04 does have a chimney which is
10.0 meters (33 feet) high, while the chimney system height in Home V10 is 6.4
meters (21 feet). It is possible that the longer chimney in Home N04 contributed
to a relatively higher creosote accumulation (due to an increased area for creosote
deposition and an increased opportunity for heat transfer frj-om the flue gas while
in the chimney). It should be noted that flue gas temperatures are recorded at the
exit of the device and not in the chimney system.
4-101
-------
Two of the three homes where Add-on J was evaluated experienced some degree of
smoke spillage. Home N04 occasionally experienced smoke spillage from the stove
air inlets and from the connection of the flue pipe adapter section which was
located between the add-on and the stove flue collar while the stove was in
operation. The study participants in Home N04 requested that Add-on J be removed.
Home N12 experienced smoke spillage from the stove air inlets and from the add-on/
flue collar connection while the stove was in operation. The study participants in
Home N12 also reported that the heat output of their conventional stove was too
high when the stove was operated at temperatures necessary to maintain catalyst
activity.
Low-emission Stoves
Stove K. Stove K was evaluated in two homes (V18 and N07). The overall mean
emission rates for these homes were 29.5 g/hr (V18) and 11.2 g/hr (N07).
The data set from Home V18 includes one sampling period (V18-7) where the measured
emission rate was the highest emission rate measured during a single sampling
period for all stoves in the study. This emission rate was accompanied by the
highest burn rate (1.26 kg/hr) measured for Stove K (the second highest measured
burn rate was 1.08 kg/hr). A review of the data set from Home V18 does not
indicate any other factors which are significantly outside the normal range
observed in this home.
The overall mean emission rates for the two homes where Stove K was evaluated
differ by 18.3 g/hr. The most significant influencing factor appears to be the
burn rate. In Home V18 the overall mean burn rate was 1.13 kg/hr (range of 1.08 to
1,26 kg/hr), while in Home N07 the overall mean burn rate was 0.86 kg/hr (two
sampling periods, burn rates of 0.84 and 0.90 kg/hr). Fueling habits also differed
in Homes V18 and N07. Home V18 had an overall mean fuel load of 3.6 kg and an
overall mean fueling frequency of 0.32 #/hr. Home N07 had an overall mean fuel
load of 5.0 kg and an overall mean fueling frequency of 0.18 #/hr.
It appears that Stove K is capable of a wide range of emission rates (9.4 to 47.6
g/hr). Furthermore, the primary factors that appear to influence emission rate for
this stove design are burn rate, fuel load size, and fueling frequency. The
limited data collected on Stove K indicate that the lowest emissions result from
4-102
-------
relatively low burn rates, relatively large fuel load sizes, and less frequent
fueling frequencies.
The relative creosote accumulations for the two homes where Stove K was evaluated
were as expected based on chimney types. Home V18 had a prefabricated metal
chimney system and creosote accumulation (one heating season) of 0.15 kg/1000 HDD.
Home N07 had an exterior masonry chimney with a round cross-section, and a mean
creosote accumulation (two heating seasons) of 1.10 kg/1000 HDD.
The study participants in Home V18 reported that it was difficult to establish
draft when lighting a cold stove. The chimney length was increased from 4.6 meters
(15 feet) to 5.5 meters (18 feet), which reportedly improved the draft
significantly when starting the stove.
The study participants in Home N07 also reported occasional difficulty in
maintaining draft while fueling the stove (the chimney length in Home N07 was 7.0
meters [23 feet]). The study participants in Home N07 reported that creosote
condensate regularly dripped from the clean-out door at the base of the chimney.
The moisture content of the fuel used in Home N07 was relatively dry (overall mean
of 20.7%). The study participants in Home N07 reported that they were very pleased
with the performance of Stove K.
Stove L. Stove L was evaluated in two homes (V04 and N15). The overall mean
emission rates for these homes were 9.2 g/hr (V04) and 9.6 g/hr (N15).
The two overall mean emission rates for the two homes where Stove L was evaluated
can be considered statistically similar. There was a relatively narrow range of
emission rates measured during the three sampling periods completed in each home
(6.5 to 14.1 g/hr for Home V04, 7.9 to 11.4 g/hr for Home N15).
There were differences in burn rates and fueling patterns for Stove L between Homes
V04 and N15, although these differences may not be large enough to be considered
significant. The mean burn rate for Home V04 was 0.90 kg/hr, while the mean burn
rate for home N15 was 1.15 kg/hr. The burn rate ranges for the two homes overlap
(0.76 to 1.07 kg/hr for Home V04, 0.93 to 1.34 kg/hr for Home N15). The mean
fueling frequency was 0.34 l/hr (range of 0.31 to 0.38 #/hr) for Home V04 and 0.45
#/hr (range of 0.32 to 0.53 #/hr) for Home N15. The overall mean fuel loads were
essentially identical for the two homes (2.7 kg for Home V04, 2.6 kg for Home N15).
The overall mean fueling frequencies for the two homes were different, but the
4-103
-------
difference may be considered insignificant given the standard deviations associated
with each mean.
A review of the data sets from Homes V04 and N15 does not indicate any significant
factors that appear to influence emission rates for Stove L. This result may
either be an indication that participate emission rates for Stove L are relatively
insensitive to operator factors, or an artifact of a relatively small data set from
two homes which operated the stove in a similar manner. It appears that Stove L
may be capable of consistently achieving relatively low emission rates, based on
the six sampling periods completed in this study, but it is difficult to make
conclusive statements based on the limited number of installations.
Home V04 had an interior masonry chimney, while Home N15 had a prefabricated metal
chimney. The relative creosote accumulations in these homes contradicts
expectations based on chimney type; Home V04 had a mean creosote accumulation (two
heating seasons) of 0.13 kg/1000 HDD, and Home N15 had a mean creosote accumulation
(two heating seasons) of 0.30 kg/1000 HDD. A review of the data set does not
indicate any factors which would explain the reversal of the expected ranking of
creosote accumulation by chimney type for the Stove L data set.
The study participants in Home V04 reported that they were very pleased with the
performance of Stove L. The study participants in Home N15 did not comment on the
Stove L performance.
Stove M. Stove M was evaluated in three homes (V12, V14, and V34). It was added
to the study for the second heating season and was selected as a stove capable of
meeting EPA 1990 emission standards. Home V12 had an emission rate (one sampling
period) of 5.2 g/hr. Home V14 had a mean overall emission rate of 21.8 g/hr (two
sampling periods). Home V34 had a mean overall emission rate of 6.9 g/hr (two
sampling periods.
The emissions performance in Homes V12 and V34 was very similar. The emission rate
in the single sampling period in Home V12 was 5.2 g/hr, and the two sampling
periods in Home V34 had emission rates of 5.9 and 7.9 g/hr. The emissions
performance in Home V14 was significantly different than in Homes V12 and V34
(emission rates of 17.2 and 26.3 g/hr for two sampling periods).
The most significant factor that differs between the Home V14 data set and the data
set for Homes V12 and V34 is the flue gas temperatures. Home V14 had a range of
4-104
-------
average flue gas temperatures of 133°C to 146°C (271°F to 295°F). Home V12 had an
average flue gas temperature (one sampling period) of 182°C (360°F). Home V34 had
a range of average flue gas temperatures of 183°C to 188°C (361°F to 370°F).
The individual burn rates, average fuel loads, and fueling frequencies for Home V14
are within the normal range of values for Stove M. Fuel moisture content ranged
between 21% and 28% for all samples. It is apparent that the stove was not fueled
significantly differently in the three homes; however, the flue gas temperatures
were generally lower and the emission rates were generally higher in Home V14. It
is difficult to formulate a hypothesis regarding the significant difference in
emissions performance of Stove M in Home V24 relative to the emissions performance
in Homes V12 and V14. The observed difference in flue gas temperatures in Home
V14, despite the similarities in operator practices in all three homes, is probably
significant. However, the data set does not appear to conclusively indicate any
factors which explain the observed difference in emissions performance in Home V14.
The relative creosote accumulations for the homes where Stove M was evaluated are
ranked as would be expected by chimney type. Home V34 had a prefabricated metal
chimney and a creosote accumulation (one heating season) of 0.22 kg/1000 HDD. Home
V12 had an interior masonry chimney and a creosote accumulation (one heating
season) of 0.72 kg/1000 HDD. Home V14 had an exterior masonry chimney and a
creosote accumulation (one heating season) of 1.05 kg/1000 HDD.
The study participants in Home V12 did not comment on the performance of Stove M.
The study participants in Home V14 reported dissatisfaction with wood heat in
general. They used a conventional stove during the 1985-86 heating season and
experienced a chimney fire. During the 1986-87 heating season (with Stove M) the
study participants reported creosote condensate dripping from the wall thimble
where the stove pipe entered the masonry chimney and ice forming at the base of the
chimney. The study participants in Home V34 did not comment on the operation of
Stove M; however, they did report a chimney fire during the 1986-87 heating season.
Stove N. Stove N was evaluated in three homes (V03, V35, and N16). It was added
to the study for the second heating season and was selected as a stove capable of
meeting EPA 1990 emission standards. The overall mean emission rate in Home V03
(two sampling periods) was 10.2 g/hr. The emission rate in Home V35 (one sampling
period) was 3.6 g/hr. The overall mean emission rate in Home N16 (three sampling
periods) was 8.2 g/hr.
4-105
-------
Emission rates which were less than 5.0 g/hr were achieved in each of the three
homes where the stove was evaluated (2.0 g/hr, sampling period V03-6; 3.6 g/hr,
sampling period V35-7; and 4.3 g/hr, sampling period N16-6). This is an indication
that the stove is capable of achieving relatively low emission rates in a variety
of installations. A review of the data sets from each home does not indicate any
significant factors which appear to have contributed to these relatively low
emission rates. The burn rate and fueling factors during these sampling periods
are not significantly different from other data sets collected in the homes.
Home V03 had higher burn rates and average fuel loads than Homes V35 and N16. The
overall mean burn rate for Home V03 was 1.33 kg/hr, compared to Home V35 (one
sampling period) at 0.90 kg/hr, and Home N16 at 0.98 kg/hr. The overall'mean fuel
load for Home V03 was 4.9 kg, while the average fuel load in Home V35 (one sampling
period) was 3.1 kg, and the overall mean fuel load in Home N16 was 3.1 kg.
There was a single emission rate recorded in Home V03 (18.3 g/hr, sampling period
V03-5) that was 8.0 g/hr higher than the second highest emission rate recorded
(10.3 g/hr, sampling period N16-7). A review of the data set from sampling period
V03-5 does not indicate any factors that appear to have significantly contributed
to the relatively high emission rate observed during this sampling period.
The data set shows that Stove N is capable of achieving low emission rates. The
data set collected on Stove N does not give a clear indication of the factors that
affect the emission rates for this stove. The low emission rates were observed
over a range of burn rates (0.90 to 1.38 kg/hr), which indicates that Stove N
emissions appear to be relatively insensitive to burn rate.
The differences between creosote accumulation in the homes are not considered
significant given uncertainties associated with the creosote accumulation
measurement method. Home V35 had a prefabricated metal chimney and a creosote
accumulation (one heating season) of 0.24 kg/1000 HDD. Home V03 had an interior
masonry chimney and a creosote accumulation (one heating season) of 0.29 kg/1000
HDD. Home N16 had an exterior masonry chimney and a creosote accumulation (one
heating season) of 0.33 kg/1000 HDD.
The study participants in Home V03 did not comment on the performance of Stove N.
The study participants in Homes V35 and N16 reported that they were very pleased
with the performance of Stove N.
4-106
-------
CONVENTIONAL STOVES ANALYSIS
Emission sampling was conducted in six homes which used six different conventional
stove models. Testing was conducted to compare conventional stoves with advanced
technology stove models and add-on/retrofits, and to establish baseline data for
existing stoves in the Northeast region.
Performance Discussion
The following discussion focuses primarily on emission rate performance and the
factors that affect emission rates in conventional stoves. As in the case of the
advanced technology stove models, the emission rates in conventional stoves can
vary significantly due to several factors, including stove design, mechanical
integrity of the stove, and operating practices.
Home V06. The conventional stove model in Home V06 had a firebox volume of 84
liters (2.95 cubic feet). This stove model is a "cabinet style" convection heater
which has the firebox surrounded by a sheet metal jacket. The combustion air
supply is designed to include both thermostatically-controlled air introduced
through two holes in the side firebirck and manually-controlled air at the fuel
loading door-
Four sampling periods were completed in Home V06. The overall mean emission rate
was 9.4 g/hr (range of 2.9 to 17.3 g/hr). This mean emission rate is the lowest
observed for all conventional stoves and lower than the mean emission rates for
several advanced technology stove models.
The emission rate for V06 is significantly influenced by two sampling periods with
relatively low emission rates during the 1985-86 heating season (V06-1, 2.9 g/hr;
and V06-2, 4.7 g/hr). During sampling period V06-1 the highest average fuel load
(8.0 kg), fueling frequency (0.31 #/hr), and burn rate (2.45 kg/hr) were measured
in this home. The average fuel load, fueling frequency, and burn rate for sampling
period V06-2 do not significantly differ from the value measured during the
remaining two sampling periods (V06-5 and V06-6) in this home.
The study participants in Home V06 consistently maintained an ash bed about 15 to
25 cm (6 to 10 inches) deep, which would serve to effectively reduce the usable
firebox size in this stove. For example, an ash bed thickness of 20 cm (8 inches)
would effectively reduce the usable firebox size in this stove to 44 liters (1.6
cubic feet), which is similar to the firebox size of many of the low-emission
4-107
-------
stoves in the study. Based on existing data on conventional stoves, a relatively
small firebox would be expected to result in lower emission rates. This
consistently high ash bed also served to completely block off the air intake holes
in the firebrick which were designed to introduce the thermostatically-controlled
primary air supply to the firebox. Without the thermostatically-controlled air,
combustion air was supplied to the firebox through an opening in the loading door
and by many leaks in the firebox. Air leaks were caused by warped metal in the
firebox, which is primarily of spot-welded sheet metal construction.
The fuel used in Home V06 occasionally included mill ends and lumber scraps, which
would be expected to have a lower fuel moisture content than cordwood. This is not
reflected in the average fuel moisture measured in this home (mean of 26.7%, range
of 25.0% to 28.0%), which is fairly normal for homes in the study.
Home V09. The conventional stove model in Home V09 had a firebox volume of 77
liters (2.7 cubic feet). The stove is a "step top" design style with a rear-exit
flue collar. The primary combustion air enters via two spin-drafts located on the
fuel loading door.
One emission sampling period was completed in Home V09 during the 1985-86 heating
season. The emission rate for this sampling period was 15.4 g/hr, which is
relatively low for conventional stoves. The most notable factor observed during
this sampling period was the high fuel moisture content (41.2%). This high fuel
moisture content did not appear to contribute to a high emission rate, although it
is difficult to determine a "typical" emission rate for this home due to the lack
of data.
Home V14. The conventional stove model in Home V14 had a firebox volume of 65
liters (2.3 cubic feet). The stove is a "step top" design style with a top-exit
flue collar. Primary combustion air is manually controlled and enters the firebox
from a slot located on the floor of the firebox near the bottom of the fuel loading
door.
Three emission sampling periods were completed in this home during the 1985-86
heating season. The overall mean emission rate was 20.2 g/hr (range of 16.9 to
23.5 g/hr).
A review of the data set does not indicate any parameters which appear to correlate
with the magnitude of individual measured emission rates. This observation may
4-108
-------
either indicate that this stove model is relatively insensitive to operation
factors or be an artifact of a small data set from a single home.
The overall mean emission rate in this home (20.2 g/hr) is essentially equivalent
to the overall mean emission rate for all conventional stoves of 20.1 g/hr.
Home N08. The conventional stove model in Home N08 had a firebox volume of 84
liters (3.0 cubic feet). The stove is a "step top" design style with a rear-exit
flue collar. Primary combustion air enters the firebox via two spin-drafts located
on the fuel loading door.
Three sampling periods were completed on this stove during the 1986-87 heating
season. The overall mean emission rate was 30.0 g/hr (range of 26.6 to 32.6 g/hr).
For the three sampling periods in this home it appears that a burn rate/emission
rate relationship exists. Caution should be used in this interpretation due to the
small data set and narrow range of data. The highest burn rate (2.19 kg/hr) was
measured with the lowest emission rate (26.6 g/hr), the middle burn rate (2.00
kg/hr) was measured with the middle emission rate (30.9 g/hr), and the lowest burn
rate (1.91 kg/hr) was measured with the highest emission rate (32.6 g/hr).
The fueling variables (average fuel load, fueling frequency, and fuel moisture
content) do not appear to exhibit any identifiable relationship to the measured
emission rates.
The overall mean fuel load in Home N08 (8.3 kg) was the largest measured for all
conventional stoves. The overall mean loading frequency (0.25 #/hr) was the lowest
measured (although equal to the overall mean loading frequency for Home V06).
Field observations indicated that the study participants in Home N08 kept the
firebox relatively full of fuel and the primary combustion air inlets at a low
setting.
Home N14. The conventional stove model in Home N14 had the largest firebox size of
all conventional stove models evaluated, 119 liters (4.2 cubic feet). The stove is
a "step top" design style with a top-exit flue collar. There are two fuel loading
doors, each with a spin-draft primary combustion air inlet.
4-109
-------
Two emissions sampling periods were completed in Home N14 during the 1986-87
heating season. The overall mean emission rate was the highest for all
conventional stoves (31.5 g/hr, two sampling periods of 29.0 and 34.0 g/hr).
The fueling frequencies and burn rates were significantly different in this home
for the two sampling periods. During sampling period N14-6 (34.0 g/hr) the fueling
frequency was 0.43 #/hr, while the burn rate was 2.45 kg/hr. During sampling
period N14-7 (29.0 g/hr) the fueling frequency was 0.29 #/hr, while the burn rate
was 1.57 kg/hr. The average fuel load size was not significantly different in the
two sampling periods, 5.6 kg for sampling period N14-6 and 5.4 kg for sampling
period N14-7. The relationship of increased emissions with increased burn rate
appears to contradict the theory for conventional stoves which says that emissions
should decrease with increased burn rate; however, the burn pattern occurring in
homes may be significantly different than the pattern obtained through laboratory
procedures on which the firebox-size-effects theory is based.
The study participants in this home occasionally operated the stove at very high
heat outputs. The stove was observed to glow red on several occasions. During
sampling period N14-6, the fueling frequency (0.43 #/hr) was the second highest
fueling frequency observed for conventional stoves. This relatively high fueling
frequency was accompanied by the highest (along with sampling period V06-1) burn
rate measured for conventional stoves (2.45 kg/hr).
Home N16. The conventional stove model in Home N16 had the smallest firebox of all
stoves in the study at 33 liters (1.2 cubic feet). This stove is a rectangular box
design. Primary combustion air enters via a circular sliding-wedge style vent
located on the fuel loading door.
One sampling period was completed in this home during the 1985-86 heating season.
This sampling period had an emission rate of 13.9 g/hr.
The single sampling period completed in Home N16 had the lowest average fuel load
(3.5 kg) and the highest fueling frequency (0.45 #/hr) of the conventional stoves.
The small firebox size, relatively small average fuel load, and relatively high
fueling frequency reflect characteristics commonly associated with the low-emission
stove classification. A modified version of this stove model has passed certifi-
cation standards (Oregon 1988 standard) as a non-catalytic stove. However, this
stove model has air inlets that direct air straight into the firebox, which is
generally not seen on low-emission stoves.
4-110
-------
The single emission rate measured for this stove is 6.2 g/hr lower than the mean
for all conventional stoves (20.1 g/hr); however, there is not sufficient data to
conclusively determine whether this emission rate is characteristic of the stove.
4-111
-------
Section 5
DISCUSSION AND CONCLUSIONS
The objective of this study was to document the performance of different types of
woodstoves as operated in typical Northeast homes. Data were collected on wood
use, creosote accumulation, and particulate emissions in 42 homes over a two-
heating-season period. Catalytic stoves, catalytic add-on/retrofit devices, non-
catalytic low-emission stoves, and "conventional" stoves were evaluated, with data
from conventional stoves serving as the baseline. One to four units of 14
different stove models or add-on/retrofit devices were installed in study homes.
The breadth of this study limited the capacity for in-depth analysis. The study is
intended to serve as a broad assessment of field performance of stoves and stove
operators.
GENERAL
The four stove technology groups (catalytic, add-on/retrofit, low-emission, and
conventional) showed consistent ranking by particulate emissions, wood use, and
creosote accumulation. While the relationships between these parameters are by no
means simple, nor the statistical significance certain in all cases, it appears
that the advanced technology devices do show improvement over conventional stoves
in all catagories. The magnitude of the improvement is affected by numerous
factors, many of which are addressed in Section 4.
WOOD USE AND CREOSOTE ACCUMULATION
Measurements of wood use were intended to provide an indication of relative
woodstove efficiency. Significant differences were observed between the stove
technology groups. While not directly correlated with measured particulate
emissions, the stove technology groups are ranked by wood use (kg/1000 HDD) in the
same order they are ranked by particulate emissions (g/hr); conventional stoves
were highest, while low-emission stoves were lowest. The lower wood use by the
advanced technology stoves and devices probably reflects both higher efficiencies
and fueling patterns characteristic of the technology.
5-1
-------
Chimney type appears to play a significant role in creosote accumulation, with
exterior masonry flues collecting the most and metal chimneys collecting the least.
This is probably due to heat losses through the chimney walls and subsequent
cooling of flue gasses. However, due to the large number of stove/chimney
combinations and limitations of the sampling method, a larger data set is needed
before conclusive statements can be made.
PARTICULATE EMISSIONS
Stove Technology Groups
Firebox size showed the strongest correlation with emission rates, and was clearly
a factor in the catalytic and conventional stove groups. Burn rate, fueling
frequency, fuel load and moisture content, catalyst operational time, and other
factors were investigated without identifying a clear relationship to particulate
emissions. The most significant observation is that stove performance data can be
highly variable, from single installations, stove models, and technology groups.
Although all measured parameters (wood use, creosote accumulation, and particulate
emissions) showed variability, particulate emissions are of special concern because
of recent EPA regulations aimed at reducing stove emissions.
Averages from stove technology groups may not be an appropriate way to evaluate
stove performance, due to several factors:
• Stoves used in the study were provided by stove manufacturers
interested in the study, and therefore do not necessarily represent
best or "typical" performance.
• Stoves were installed in homes without any detailed verbal
instructions given to homeowners on the use of their new stove.
Although they were provided with the stove instruction manual, it is
possible that if homeowners were to purchase the stove, more time
would be spent on user education.
• The study was conducted in areas of New York and Vermont which
average about 8,000 to 9,000 heating degree-days (Fahrenheit basis)
per year. Stoves are burned at higher rates than other regions,
which may increase emissions from catalytic stoves and add-ons/
retrofits and reduce emissions from non-catalytic and conventional
stoves.
• Although "Student's t" test results show that the data sets are
probably different, it is unclear how different the values would be
if the same stoves were used under different conditions. It should
be stressed that these results reflect specific stoves in specific
installations, operated and fueled in a specific manner.
• Stove and catalyst technologies were not equally represented. Stoves in
the catalytic, add-on/retrofit, and low-emission categories included
5-2
-------
models certified to Oregon DEQ 1986 and 1988 standards and EPA 1990
standards, as well as non-certified stoves. A variety of combustor types
and thicknesses were used. Some combustors were replaced in mid-study.
Stove Models
For the reasons mentioned above, stove performance is best evaluated by examining
individual installations. Several stoves appear to work well in one installation,
but poorly in another, indicating that while the stove may be capable of low-
emission performance, other factors can be significant. In some cases there are
major differences in stove performance in a given home during sampling periods.
Overall, there 'did not appear to be a progressive increase in emissions over the
two-heat ing-season period.
Stove D, which had new combustors installed at the start of the 1986-87 (second)
heating season, showed marked reduction in emission rates. However, due to the
problem with deteriorating combustors noted during the first heating season, it is
not clear whether the reduced emissions were due to better actual catalytic
performance or less operating time with deteriorating combustors. In other words,
the apparent improvement in performance may be due to the stove operating more as a
catalytic and less as a non-catalytic. Stove D had the lowest average emissions in
the catalytic stove group. It should be noted, however, that the low emission rate
reflects relatively frequent stove inspections and the replacement of combustors.
Without stove inspections, emissions would likely have been higher.
It may be significant that among the catalytic stoves, average stove emissions are
ranked by firebox size (Retrofit E is an apparent exception). Large firebox
stoves, when not operating catalytically, may produce higher emissions, increasing
average overall emissions. The integrated one-week samples appear to represent
significant periods of non-catalytic operation, as documented in Table 3-10A. If
emissions are higher during non-catalytic periods from large firebox stoves,
overall average emissions would be expected to be higher. Stove D, with the lowest
emissions among catalytic stoves, had the smallest firebox. However, each stove
had at least one catalyst replacement during the two-year study.
The variability of emissions from a given stove model between homes suggests that
caution should be used when evaluating stoves in the field. With two or three
installations per stove model, it is difficult to tell whether measured emissions
are representative. Consistent emissions from a single home may simply reflect
consistent operation practices by the homeowner. Considering the range of values
5-3
-------
measured, data sets should be larger before conclusions can be made with a high
degree of confidence.
Most stove models, existing or provided for the study, had relatively low emissions
for some periods in some homes. This includes conventional stoves. The indication
is that operational and fueling practices can significantly reduce particulate
emissions. Virtually all stove models with small fireboxes (with the exception of
one Stove K home) had relatively low emissions. The small firebox sizes found in
three stoves may act as a "governor," limiting maximum emissions when the stoves'
low-emissions features are not active. The limitation may be in the form of
enhanced combustion, smaller fuel loads, or more frequent "burn down" phases.
While certainly not the only factors in explaining stove performance, these may be
significant ones.
The apparent low-emission/small firebox size relationship may reflect the
parameters used to define "stove operational time." For this study, a stove was
considered operational if flue gas temperatures at the exit of the appliance were
greater than 38°C (100°F). This value may be low enough to include long periods of
"charcoal phase" burning when particulate emissions would be low. Review of
temperature data (see Appendix D in Volume II, a companion document to this report)
from the sampling periods indicates that smaller firebox stoves tend to burn down
more frequently before refueling, which may result in more sampling during charcoal
phase periods.
It is important to note that emission samples represent one-week averages, during
which time an average of 30 to 50 fuel loads are added. Stoves with high average
emissions may have short but acute periods of high emissions which raise the
overall average.
Many of the parameters investigated (burn rate, fueling practices, alternative
heating system use) did not appear to correlate well with particulate emissions,
although general trends appeared in some cases. The small data sets, the large
degree of variability, and the number of potential variables made more detailed
analysis difficult.
Significant findings from emission testing in study homes include:
1.0 Advanced Technology Performance
1.1 Most stoves in the advanced technology categories (catalytic, add-
on/retrofit, low-emission non-catalytic) episodically demonstrated
5-4
-------
lower emissions than the baseline conventional stoves under "field
use" conditions. Good performance in at least one installation for
most of the stove models indicates that factors, such as stove
maintenance and fueling practices, may be as important as stove
technology features in achieving low emission rates. Stove firebox
size, regardless of technology group, was a prime factor in
determining emission rates; smaller stoves had lower emissions.
1.2 In general, performance of the stove technology groups appeared to
be consistently ranked in terms of particulate emission rates, wood
use, and creosote accumulation; low-emission non-catalytic stoves
had the lowest particulate emission rate, wood use, and creosote
accumulation, while conventional stoves had the highest. It should
be noted that only low-emission non-catalytic stoves showed a mean
emission rate which was statistically different from the
conventional stoves. It should also be noted that creosote
accumulation is strongly influenced by flue system type and wood use
appears to be influenced by burning patterns and firebox size.
1.3 All advanced technology stove groups averaged lower wood use and
creosote accumulation rates when households switched from
conventional stoves between heating seasons. Average reductions by
stove group ranged from about 10% to 35% for creosote and from about
15% to 30% for wood use.
1.4 The low-emission stoves, as a group, had the lowest average
emissions. Each model had different burning characteristics; most
showed relatively good performance. Average results from this
technology group are strongly influenced by the good performance of
two EPA 1990-certifiable stoves (M and N). Furthermore, excluding
one high-emission home (V18, using non-EPA-certified Stove K) would
reduce average emissions in this category from 13.4 to 10.0 g/hr,
and reduce the standard deviation (a-) from 10.2 to 5.7-
1.5 User satisfaction was generally high with the advanced technology
stoves provided to study homes. In particular, homeowners with
catalytic and low-emission stove models were frequently pleased with
the units. (In some cases, user satisfaction remained high even
though the catalytic combustor had deteriorated.) Some add-on
devices also received positive comments. The add-on with the lowest
average particulate emission rate also received homeowner complaints
about smoke spillage.
2.0 Catalyst Performance
2.1 Catalytic stoves showed variable performance. Most individual
models performed well in some homes. Other installations had
relatively high emissions. Overall, performance of these stoves did
not match the expectations created under ideal laboratory
conditions, although only one of the catalytic models was EPA 1990
certifiable. The mean emission rates of existing catalytic stoves
and new catalytic stoves were virtually identical. User education
and further technology refinements remain possible factors which
could help improve the performance of catalytic stoves.
2.2 Add-on/retrofit devices did not perform well overall, but 2 devices
reduced emissions considerably. The stoves on which these devices
were mounted are a major factor in measured emission rates.
5-5
-------
Retrofit F, which consistently had high emissions, is no longer
being produced.
2.3 Catalyst durability was quite variable. Rapid deterioration was
noted in some combustors, all of which were cordierite-based, with
corresponding increases in emissions. In one stove model (which
apparently accelerated combustor deterioration), replacement with
"second generation," non-cordierite combustors appeared to virtually
eliminate the deterioration trend. Emissions from this stove model
were reduced by about one-third by using "second generation"
combustors during the second year, although it is not clear whether
this was from improved catalytic performance or reduced degradation.
2.4 While a number of catalytic stoves showed increased emissions over
the course of the study, others did not. No distinct trend of long-
term loss of effectiveness was noted. However, a number of
combustors (cordierite-based) were discovered to be deteriorating.
These combustors were replaced; emission values reported in this
study reflect relatively frequent catalyst inspections and
replacement when necessary. It should be noted, however, that not
all cordierite-based combustors in the study indicated signs of
deterioration of the substrate. A cordierite-based combustor from
an "existing" stove with an estimated 6000 hours of use showed
relatively low emissions in lab retesting. All combustors retested
in the laboratory had reduced performance relative to new
combustors.
2.5 Condensation of moisture and organic material in flue systems and
subsequent drainage or leaching of condensate was a problem in some
homes during periods of very cold (< 20°C) weather. Only catalytic
stoves experienced this problem. This appears to be related to
inappropriate installation and is not necessarily a technology
1 imitation.
2.6 Catalyst AT (temperature change across the combustor) and %
operation time are not good indicators of stove particulate
emissions. Factors such as fueling cycles (long burn-down "tails")
and measurement difficulties may preclude the use of these
parameters for predicting emission rates.
3.0 Operator Practices
3.1 Operator practices, in combination with other parameters, appear to
be a significant factor in stove performance. Specific practices
which may result in lower emissions from all stoves have not been
identified from available data. However, routine maintenance
inspections of the combustor, gasketing, and overall stove system
can help identify deteriorated components in need of repair or
replacement.
3.2 Burn rates did not demonstrate a strong correlation with emission
rates for any of the stove technology groups, although "general
trends" were observed. Often, as in the case with conventional
stoves, the trend was opposite that which was expected; emissions
increased with burn rate. This may be related to field conditions,
in which lower burn rates may include longer "charcoal phase"
burning periods.
5-6
-------
3.3 Mean fuel loading frequencies were identical for the low-emission
and conventional stove groups, although the average low-emission
stove fuel load was 56% that of the average conventional stove fuel
load. This indicates that smaller firebox capacity (typically
associated with low-emission stoves) does not necessarily require
more frequent fueling of the stove. User satisfaction was generally
high with the low-emission stoves.
3.4 Average emission factors (g/kg) for all the stove catagories were
quite similar. Differences in average emission rates (g/hr) were
therefore driven by burn rates. The low average burn rate of the
low-emission stoves, and resulting low average emission rate, may be
due to more frequent "charcoal phase" burning periods.
3.5 Fuel loading frequencies did not correlate well with particulate
emissions. However, loading frequencies did increase with smaller
fuel loads for all technology groups, as was expected.
3.6 Fuel loading frequencies were significantly different between homes,
even those using the same stove model.
3.7 The lack of strong correlations between particulate emissions and
other variables indicated that many parameters have significant, if
unquantified, effects on stove performance. Fueling and burning
cycles are thought to be areas for further investigation.
4.0 Technology Factors
4.1 Firebox size is a major factor in determining particulate emissions
from woodstoves; emission rates increased with firebox volume,
regardless of stove technology.
4.2 Preliminary results from stove inspections conducted after the
second heating season (September 1987) indicate that significant
"leakage" of smoke around combustors may be a cause of high
emissions in some stoves. (A report on this work will be issued
under separate cover.) Stove inspections showed that gasketing,
especially around the bypass damper and combustor, was the most
frequent component in need of maintenance and the apparent cause of
leakage. Leakage rates and particulate emissions do not appear to
correlate well overall, but show some correlation for individual
stove models.
4.3 Using a qualitative measurement methodology, insulated metal chimney
systems accumulated the least amount of creosote. Masonry chimneys
located on outside walls accumulated the most.
5.0 Other Findings
5.1 This study did not show that one stove model is necessarily "better"
than another. As stated previously, a wide range of results were
recorded. For a given stove model, the largest number of emission
samples was 19; the smallest was 1. The largest number of instal-
lations for a given stove model was 4, while the smallest was 1.
The high degree of variability in performance and the relatively
small sample populations make comparisons inappropriate.
5-7
-------
5.2 Conventional stoves in this study may be cleaner-burning heaters
than are "typical." Four of the six conventional stoves had
relatively small fireboxes (< 2.4 ft^), and two of these had small
effective fireboxes (< 1.5 ft^). Emissions from these stoves
therefore may not be typical of existing stove technology.
Additionally, the cold Northeast climate and commensurately higher
burn rates preclude direct comparison to stove performance in milder
climates.
5.3 Alternate heating system use did not correlate well with particulate
emission rates or burn rates, although heating system use was
monitored only in the room with the stove. In general, most homes
in the study used their alternate heating system less than 3.5% of
the time (while the stove was operating). This amounts to less than
one hour per day. A large portion of the homes used no back-up heat
at all.
5.4 Polycyclic organic material (POM) emissions were variable and non-
conclusive. Testing method and analytical method limitations, and a
very limited database, preclude any ranking of POM emissions by
stove type.
5-8
-------
Section 6
RECOMMENDATIONS
Many factors can affect woodstove performance, as evidenced by the wide range of
values from individual stove models and stove technology groups. Identifying what
factors are most significant in causing high or low emissions is difficult due to
the large number of variables and the relatively small sample populations.
Continued investigation of woodstove performance under "field use" conditions will
be likely in future studies, and this section identifies key areas which should
receive additional attention.
DATA REDUCTION/EXISTING DATA BASE
Detailed Graphics
Many of the stoves show wide ranges of catalyst temperatures and flue gas
temperatures and oxygen content. Wide swings in these values are usually
associated with fueling events.
It is suspected, based on data review and from observations in the laboratory, that
particulate emissions can be quite high immediately after fuel is placed in the
stove or when the stove is started cold. Graphs of stove temperatures, flue gas 02
content, fueling weights and frequencies, and auxiliary heating system use are
presented in Appendix D. These graphs show data from the entire sampling period. A
more detailed review of data would allow investigation of stove and catalyst
temperatures and flue gas 02 immediately following each fueling episode. This
could be done by expanding the scale of the graphs to allow review of data on a
day-by-day basis. Differences between "high emission" and "low emission" samples
could then be identified based on how the stove responds to fuel loading events.
Review of Field Studies
Several other field studies have been conducted in the past two years. Sampling
methodologies are similar, while climate, fuel types, and other factors are
different. Some stove types are identical. Review of all available data may allow
additional insight into factors affecting stove performance.
6-1
-------
Evaluation of Stove Design Factors
Stove design features which help maximize stove performance should be identified.
This could be investigated by evaluating stove emission data, home inspection
reports, and lab testing. Features which are prone to failure should also be
identified.
ADDITIONAL FIELD STUDY
Stove Inspections
Stoves in all homes where valid emission samples were collected were inspected in
September 1987. Combustors were replaced in all catalytic stoves and devices, with
the used combustors archived for future testing and analysis. During stove
inspections, a leak-testing device was used to measure significant bypass flow (if
any) around combustors. This will provide a potential explanation of high
emissions from some catalytic stoves. A report on the inspections and leak testing
will be prepared in a separate report.
Additional Stove Testing
Field Testing. Selected stoves should be tested under field use conditions with
detailed monitoring before, during, and after testing. Instrumentation should be
used to document stove door and bypass lever usage. AWES samplers could be used in
tandem, operating seguentially during different periods (high/low temperatures,
before/after fuel loading, etc.). Data LOG'rs should be modified to record
temperatures every five minutes. Switching stoves of a specific model from high
emission installations to homes which previously showed low emissions could help
identify whether technology or operator factors were primarily responsible for
stove performance.
A smaller sample size in a single area (New York or Vermont) would allow more
samples in a given stove installation. Study stoves should be purchased and given
to homeowners, or a commensurate cash payment made.
Laboratory Testing. Factors which are identified as potentially significant from
review of field use data could be applied under controlled laboratory conditions.
Effects would be documented under more controlled conditions than field testing.
Size of fuel loads, fueling frequency, burning patterns, stove maintenance, and
other factors are potential variables to be investigated.
6-2
-------
Combustor Testing. The emission reduction potential of catalytic combustors could
be verified in laboratory testing, isolating one variable under controlled
conditions. All combustors from the second heating season of the study will be
archived and will be available for testing. A combination of bench testing (using
gas fuel or wood smoke) and stove testing (using a standard stove and fuel) would
document combustor longevity. Combustors showing very high or very low emissions
from bench testing should be selected for additional testing in a standard stove.
If the combustor is shown to have lost catalytic activity, additional testing to
determine the cause of the loss (exposure to high temperatures, "poisoning" by
various compounds, etc.) should be conducted.
Stove Selection. Based on the results of testing and analysis conducted to date, a
series of tests could be devised to identify stoves which should work well under
field conditions. Stoves which pass these lab screening tests would be installed
in field homes for future testing. Stoves with design factors which were shown to
ensure low emissions in the field could then be encouraged.
6-3
-------
Section 7
REFERENCES
1. Oregon State University Extension Service, Extension Circular 1023, September
1980, Corvallis, Oregon.
2. Standards of Performance for New Stationary Sources, Standards of Performance
for New Sources, Residential Wood Heaters; Listing of Residential Wood Heaters
for Development of New Source Performance Standards; Proposed Rules, Federal
Register, February 18, 1987, pages 4994-5066.
3. GC/MS—Modified EPA Method 625, Federal Register, October 26, 1984, pp. 43385-
43406.
4. Truesdale, R. S., et al., "Final Report: Characterization of Emissions from
the Combustion of Wood and Alternative Fuels in a Residential Woodstove,"
EPA--600/7-84-094, September 1984, NTIS No. PB85-105336.
5. Burnet, P. G., and Tiegs, P. E.; "Woodstove Emissions as a Function of Firebox
Size"; presented at the 1985 Wood Heating Alliance Technical Seminar,
Baltimore, March 1985.
7-1
-------
Appendix A
Study Home Characteristics
-------
Table A-l
NCS STUDY HOME CHARACTERISTICS
VERMONT HOMES
ID
V01
V02
V03
V04
V05
V06
V07
V08
V09
V10
Vll
V12
V13
V14
V15
V16
V17
V18
V19
V20
V21
V22
V23
V24
V25
V26
V27
V28
V29
V30
V31
V32
V33
V34
V35
WOODSTOVE9/
85/86
A/R(E) *
A/R(G) *
A/R/(F) *
L.E.(L) *
Cat.(B) *
Conv.(O) *
Cat.(C) *
Cat.(D) *
Conv.(O) *
A/R(H) *
Cat.(B) *
A/R(F) *
Cat.(D) *
Conv.(O) *
A/R(H) *
Cat.(C) *
Cat. (P)
Cat.(D)
_^_
Conv.(O)
Conv..(0)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Cat.(C)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Cat.(P) *
Cat.(P) *
Cat.(P) *
—
—
86/87
A/R(E) *
A/R(S)
L.E.(N) *
L-E.(L) *
Cat.(B) *
Conv.(O) *
Cat.(C) *
Cat.(D) *
—
A/R(J) *
Cat.(B) *
L.E.(M) *
Cat.(D) *
L.E.(M) *
Conv.(O)
Cat.(C) *
Cat.(P)
L.E.(K) *
Cat.(C)
Cat.(B)
A/R(6)
Cat. (A)
L.E.(K)
A/R(I)
—
Conv.(O)
Conv.(O)
Cat. (A)
A/R(E)
A/R(I)
Cat.(P) *
Cat.(P) *
Cat.(P)
L.E.(M) *
L.E.(N) *
CHIMNEY
TYPEb/
VII
VIII
VI
V
VIII
V
VII
VI
VI
VII
V/XII
VI
VI
VIII
VIII
VII
II
V
I
V
I
III
V
V
VII
I
IV
I
IV
V
VIII
XI
VI
IX
I
I
ALTERNATE HEATING0/
TYPE
gas
none
gas
oil
gas
electric
oil
oil
electric
oil
electric
oil
oil
oil
electric
oil
none
gas
electric
electric
electric
electric
kerosene
gas
electric
none
electric
electric
oil
oil
oil
electric
electric
oil
gas
electric
FREQUENCY OF USE
never
never
frequently
frequently
occasionally
never
occasional ly
rarely
rarely
occasionally
rarely
occasionally
rarely
frequently
occasional ly
rarely
never
frequently
occasionally
occasional ly
never
daily
occasionally
rarely
rarely
never
rarely
rarely
rarely
rarely
rarely
frequently
rarely
rarely
frequently
rarely
A-l
-------
Table A-l (Continued)
NEW YORK HOMES
ID
N01
N02
N03
N04
N05
N06
N07
N08
N09
N10
Nil
N12
N13
N14
N15
N16
N17
N18
N19
N20
N21
N22
N23
N24
N25
N26
N27
N28
N29
N30
N31
N32
N33
WOODSTOVE3/
85/86
Cat. (A) *
Cat.(D) *
Cat.(C) *
A/R(G) *
A/R(F) *
A/R(I) *
L.E.(K) *
Conv.(O) *
Cat.(B) *
Cat. (A) *
Cat.(D) *
A/R/(I) *
A/R(H) *
A/R(I) *
L.E.(L) *
Conv.(O) *
Conv.(O)
Conv.(O)
Cat.(C)
Conv.(O)
Conv.(O)
Conv.(O)
Cat.(B)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(O)
Conv.(P)
Cat.(P) *
Cat. (A) *
Cat.(P) *
86/87
Cat. (A) *
Cat.(D) *
Cat.(C) *
A/R(J) *
A/R(G) *
Conv.(O) *
—
L.E.(K) *
Conv.(O) *
Cat.(B) *
Cat. (A) *
Cat.(D) *
A/R(J) *
Conv.(O) *
L.E.(M) *
A/R(J) *
Conv.(O) *
L.E.(L) *
L.E.(N) *
Conv.(O)
Cat.(B) *
—
Cat. (A)
Conv.(O)
Cat.(B)
—
A/R(E)
Conv.(O)
A/R(E)
A/R(G)
—
L.E.(K)
L.E.(K)
Cat.(P)
Cat. (A) *
Cat.(P) *
CHIMNEY
TYPE5/
I
VIII
V
VII
VIII
VII
XIII
VII
VII
VIII
V
I
VII
VII
II
VIII
IX
VII
II
VII
V
I
VII
I
X
VI
VII
V
V
VII
VIII
I
XII
ALTERNATE HEATINGC/
TYPE
gas
oil
solar
oil
gas
gas
gas
oil
gas
electric
electric
oil
electric
oil
gas
electric
oil
electric
gas
electric
electric
electric
electric
electric
oil
electric
gas
electric
electric
gas
none
electric
gas
FREQUENCY OF USE
rarely
rarely
frequently
occasional ly
frequently
occasional ly
occasional ly
occasional ly
occasional ly
dai ly
occasional ly
frequently
rarely
rarely
daily
frequently
occasionally
daily
rarely
occasional ly
rarely
rarely
dai ly
never
dai ly
never
rarely
rarely
never
rarely
never
rarely
never
A-2
-------
Table A-l (Continued)
a/ Woodstove Identification: The letters "A" through "P" refer to the type of
woodstove being used in each heating season. These stove code letters are
preceded by the technology category each stove falls into. The woodstove types
followed by an asterisk (*) indicate those study homes in which AWES emissions
sampling was conducted for each heating season. The study stoves fit into the
following categories:
A-D Integrated catalytic woodstoves provided by the study;
E-F Retrofit catalytic woodstoves;
G-J Catalytic add-on units;
K-N Low-emission non-catalytic woodstoves;
0 Conventional old-technology woodstoves;
P Existing catalytic woodstoves in use before the start of the study.
b/ Chimney Type: The Roman numerals "I" through "XIII" refer to the type of
chimney in use with the study stove. Chimney identifications are as follows:
I. Prefabricated metal chimney with no bends and a straight-up installation
through the ceiling(s) and roof;
II. Prefabricated metal chimney with two ninety-degree bends so as to pass
the chimney through a wall to the exterior of the house;
III. Prefabricated metal chimney exiting up through the ceiling(s) and roof,
but with two ninety-degree bends to offset the stove from the chimney
location;
IV. Stainless-steel-lined masonry chimney located inside the exterior walls
of the house;
V. Square-tile-lined masonry chinney located inside the exterior walls of
the house (cross-sectional area approximately 7 inches by 7 inches);
VI. Rectangular-tile-lined masonry chimney located inside the exterior walls
of the house (cross-sectional area approximately 7 inches by 11 inches);
VII. Square-tile-lined masonry chimney located outside the exterior walls of
the house (cross-sectional area approximately 7 inches by 7 inches);
VIII. Rectangular-tile-lined masonry chimney located outside the exterior
walls of the house (cross-sectional area approximately 7 inches by
11 inches);
IX. Stove vents into a fireplace with a rectangular-tile-lined masonry
chimney located outside the exterior walls of the house (cross-sectional
area approximately 7 inches by 11 inches);
X. Stove vents into a fireplace with a large square-tile-lined masonry
chimney located outside the exterior walls of the house (cross-sectional
area approximately 12 inches by 12 inches);
XI. Stove vents into a prefabricated zero-clearance fireplace with a
prefabricated metal chimney (6-inch diameter) located inside the
exterior walls of the house;
XII. Round-tile-lined masonry chimney located inside the exterior walls of
the house (cross-sectional area 8 inches in diameter);
XIII. Round-tile-lined masonry chimney located outside the exterior walls of
the house (cross-sectional area 8 inches in diameter).
c/ Frequency of use reported is based on the homeowners' estimated use. It does
not distinguish between how often the alternate heating system was used and how
many hours it was used, however. Refer to Table 3-10A in the text for computer-
documented use of alternate heating system use.
A-3
-------
Table A-2
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIES"/ FREQ. OF USE0/ NOTES
V01
V02
A/R (E)
for all
sampl ing
runs
A/R (G)
for all
sampl ing
runs
Ext.
7" x
21 ft
Ext.
7" x
(VIII
22 ft
masonry,
7", (VII)
. high
masonry,
H",
),
. high
1370 /
748/
1 +
basement
1426/
1426/
2 +
basement
Forced air gas
furnace: 1.1%
usage.
Woodstove is
only heat
source.
Family
Two
1/87.
Fami ly
Home
of four
catalyst
of four
; both spouses
substrate fai
; both spouses
occupants report they
with add-on's
performance.
work
lures
work
were
days.
: 3/86
days.
and
pleased
-------
Table A-2 (Continued - Page 2)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTQVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY9/ #QF STQRIESb/ FREQ. OF USEC/ NOTES
V03
A/R (F)
for
samples
1-3,
I.E. (N)
for
samples
5-7
Int. masonry, 1362/
7" x 11", (VI), 1482/
22 ft. high 2 +
basement
Forced air gas
furnace: 1.3%
usage.
Wood cook-
stove: used
less than five
times during
study for
cooking and
extreme cold
periods.
Electricity:
baseboard heat
in office area
addition of
home (240 sq.
ft.); used
daytime only
approx. 1/3 of
days.
Working couple; husband runs business at
home and in separate barn.
Retrofit stove produced smoke intrusion
into home on at least four occasions while
stove was left unattended. Smoke exited
stove through secondary air inlet port of
add-on/retrofit.
Retrofit stove was used with a hot water
loop plumbed through the lower foot of flue
pipe. This system was pulled aside (but
touching the exterior of the flue pipe) when
the low emission stove requiring a 6" flue
was installed.
Low emission stove's insulation support
brackets (inside firebox) warped and were
eventually removed and replaced.
Replacement brackets were supplied by
manufacturer and installed by OMNI field
personnel.
-------
Table A-2 (Continued - Page 3)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIES'3/ FREQ. OF USEC/ NOTES
V04
L.E. (L)
for all
sampling
runs
Int. masonry,
7" x 7", (V),
13 ft. high
1282/
Approx.
2000/
2 +
basement
Oil forced air
furnace: 1.1%
usage during
sample periods
in family room
with stove.
Working couple, but often someone home
during part of day.
Homeowner reports frequent use of oil
furnace heating system. Family room furnace
vent was often closed off with a rug, and
other areas in home may require oil heat
more often than family room.
Home was undergoing major remodeling
during study period, including changes in
floor space and weatherization of home.
Exceptionally dry firewood supply; most
wood seasoned over two years and stored in
heated areas of home.
Home occupants report they were very
pleased with stove performance.
-------
Table A-2 (Continued - Page 4)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY9/ #OF STORIES"/ FREQ. OF USEC/ NOTES
V05
Cat. (B)
for all
sampling
runs
Ext. masonry,
7" x 11",
(VIII),
22 ft. high
(85/86)
24 ft. high
(86/87)
1710/
1710/
1 +
basement
Gas forced air
furnace: 1.9%
usage
Family of four; both spouses work days.
Major flue condensation/chimney icing
problems resulted from using this catalytic
stove model. Creosote and condensate
leached into the masonry blocks along the
entire length of chimney system. Lower
chimney ice formation occurred regularly
during extended cold periods. Homeowner
reports attempts to reduce condensation by
running stove hotter or in bypass longer
with limited success. Insulation of chimney
and extending chimney height were also done
during study in attempt to reduce the
problem. Heat exchange baffles inside stove
were removed after sample rotation 6 in
attempt to reduce flue condensation.
Catalytic combustor was misaligned for
sampling rotations 1-3 in first study year.
Misalignment of combustor in stove is
thought to have happened during
installation.
Experienced chimney fire 3/86.
Firewood is split into smaller than
average pieces.
Homeowner is upset at the damage his
chimney and home have undergone since using
this study catalytic stove and has demanded
the situation be resolved.
-------
Table A-2 (Continued - Page 5)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIES"/ FREQ. OF USEC/ NOTES
V06 Conv. (0)
for all
sampling
runs
I
Co
V07 Cat. (C)
for all
sampling
runs
Int. masonry,
7" x 7", (V),
27 ft. high
1911 /
1911 /
2
Ext. masonry,
7" x 7", (VII),
21 ft. high
1720/
860/
1 +
basement
Zoned electric
baseboard
heat: 0.0%
usage in
living room
zone. No use
in other zones
reported.
Oi1 hot water
furnace: 27.9%
usage.
Family of four; both spouses work days.
Large south-facing windows contribute
passive solar heat.
Homeowner maintains unusually high ash bed
inside stove (6-10" high).
Firewood is cut to 24" lengths.
Homeowner occasionally burns considerable
quantities of mill ends and lumber scraps.
Family of four; both spouses work days.
Firewood is split into larger-than-average
size pieces. Firewood is stored in
basement, which is kept very warm.
Stove occasionally operated extremely hot
with cherry red surfaces (while in bypass
mode).
Stove heats basement of home and only a
small vent serves to direct heat upstairs,
producing a much warmer basement to achieve
desired temperatures in upstairs living
area.
Homeowner reports dissatisfaction with
heat output of stove.
Home occupants report they will not heat
with wood after the end of this study.
Reasons cited are primarily due to the
nuisance involved. Interest in wood heating
waned during study and homeowner was
eventually burning wood just for the benefit
of the study.
-------
Table A-2 (Continued - Page 6)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE HOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #QF STORIESb/ FREQ. OF USEC/ NOTES
V08 Cat. (D) Int. masonry, 864/ Oil forced
7" x 11", (VI), 864/ furnace: 0.
18 ft. high 1.5 + usage.
basement
air Family of three; both spouses work days.
•0% Child was born between 85/86 and 86/87
heating seasons.
Oil furnace in basement vents to same flue
as woodstove.
Homeowner reported three chimney fires
during study: 2/86, 11/86, and 11/86.
Catalyst substrate failure noted 4/86.
V09 Conv. (0)
Int. masonry,
7" x 11", (VI),
14 ft. high
851/
851/
1 +
basement
Zoned electric
baseboard
heating: 0.0%
usage. No use
in other zones
reported.
Second wood-
stove on the
same flue was
added to
basement 3/86
and used regu-
larly. This
was cause for
dropping stove
from study.
Three adults live in home; patterns of
presence at home were variable.
Firewood was poorly seasoned, usually wet
and often rotted wood.
-------
Table A-2 (Continued - Page 7)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
ffOF STORIESb/ FREQ. OF USEC/ NOTES
V10
I
t—'
CD
A/R (H)
for
samples
1-3,
A/R (J)
for
samples
4-7
Ext. masonry, 1685/
7" x 7", (VII), 1685/
21 ft. high. 2 +
Unusual config- basement
uration using
four bends
Oil hot water
furnace: 5.2%
usage.
Single homeowner working days and often away
from home in evenings. Occasionally
operates bed and breakfast style
accommodations in home.
Firewood varies greatly in size and
length, with a significant amount of small
diameter wood.
Experienced ash plugging and smoke
spillage into home with add-on H.
Experienced smoke spillage into home with
Add-on J following initial installation of
add-on. Correction of installation error
eliminated smoking problem.
Homeowner was using add-on J in its
partial bypass mode only until the start of
sample run 6.
-------
Table A-2 (Continued - Page 8)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ IOF STORIES"/ FREQ. OF USE0/ NOTES
Vll
Cat. (B)
for all
sampling
runs
85/86:
Int. masonry,
7" x 7", (V),
25 ft. high
86/87:
Int. masonry,
8" dia., (XII),
25 ft. high
1102/
1102/
2 +
basement
Zoned
electrical
baseboard
heat: 0.0%
usage. No use
reported in
other zones.
Single homeowner working days and often away
from home in evenings.
Experienced flue condensation problems for
much of study period, including condensate
and creosote leaching into the masonry
blocks of the chimney and excess water
collecting in base of chimney. Chimney was
rebuilt with the recommended flue size in
11/86, but water collection in chimney base
persisted. Homeowner made several calls to
manufacturer about the problem and attempted
to run stove hotter and in bypass position
longer at manufacturer's request. This
provided limited success in solving problem.
Heat exchange baffles inside stove were also
removed during samples 2 and 3 in attempt to
reduce problem, but were reinstalled before
86/87 sampling began.
Homeowner noted a change in stove's air
control thermostat during the second study
year with often erratic stove operation.
Homeowner reports her involvement in the
study and use of this catalytic stove was
more trouble and expense that she "ever
dreamed possible."
-------
Table A-2 (Continued - Page 9)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE
V12
V13
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
WOODSTOVE CHIMNEY3/ #OF STORIESb/ FREQ. OF USEC/
A/R (F)
for
samples
1-3,
L.E. (M)
for
samples
4-7
Cat. (D)
for all
samples
Int. masonry,
7" x 11", (VI),
22 ft. high.
Configuration
uses three
bends.
Int. masonry,
7" x 11", (VI),
20 ft. high
1377 /
1666/
2 +
basement
13817
19557
2 +
basement
Oi 1 hot water
furnace: 0.0%
usage on
radiator
monitored
during sample
periods.
Occasional use
of furnace for
other areas of
house was
reported.
Oil forced air
furnace: 0.2%
usage.
NOTES
Working couple; both spouses work days.
Oil furnace in basement vents to same flue
as woodstove.
Working couple, but often someone home days.
Oil furnace and wood furnace in basement
vent to same flue as woodstove.
Experienced two catalyst substrate
Wood furnace:
reported use
of less than
five times per
heating
season.
failures: 2/86 and 4/86.
Stove was observed to have ash pan access
door open for much of its operating time.
Homeowner used a non-catalytic model of
this same stove prior to the start of the
study.
Several large bricks of soapstone were
placed on top surface of stove while
operating in this home.
Home occupants report they were pleased
with this catalytic stove's performance,
especially the reduction in creosote.
-------
Table A-2 (Continued - Page 10)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY5/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#QF STQRIESb/ FREQ. OF USE0/ NOTES
V14
Conv.(O)
for
samples
1-3,
L.E. (M)
for
samples
4-7
Ext. masonry,
7" x 11",
(VII),
21 ft. high
9887
9887
2+
basement
Oil forced air Family of four; one spouse works days.
furnace: 1.2%
usage
(average)
Experienced chimney fire 3/86.
Home occupants report they have had a
history of chimney fires prior to study
start (approximately one per year).
Majority of firewood was cut in half to
produce pieces 8-10 inches long to fit
inside smaller firebox of Stoves 0 and M
(Stove 0 was provided to homeowner as part
of study).
Experienced flue condensation and lower
chimney icing with low emission stove M.
Main flue condensation problem was creosote
dripping into home from stove pipe/chimney
connection. Ice would form in base of
chimney during cold weather periods.
Homeowner reports attempts to modify burning
habits to reduce flue condensation were not
successful.
Homeowner reports dissatisfaction with
both woodstoves used during study due to
high creosote buildup, chimney fires, and
flue condensation.
Home occupants report they will not be
heating with wood beyond the completion of
this study due to problems they have
experienced with woodstoves and competitive
pricing of fuel oil.
-------
Table A-2 (Continued - Page 11)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE UOODSTQVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY9/ IOF STORIES"/ FREQ. OF USEC/ NOTES
V15 A/R (H)
for
sample
runs 1-3
Ext. masonry, 1650/
7" x 11", 1650/
(VIII), 2
24 ft. high
Zoned electric
baseboard
heat: 0.0%
usage in
fami ly room
during
sampling.
Other zones
are used
occasional ly.
Family of five; one spouse works days.
Experienced smoke intrusion into home
using this add-on.
Homeowner reports opening windows near
stove regularly to help draft, air out room,
or release excess heat.
Homeowner reports kids often turn up
thermostats for electric heat in bedrooms
and parents keep turning them off.
Homeowner reports dissatisfaction with
add-on performance and requested the add-on
be removed as soon as enough data was
obtained using it.
V16
Cat. (C)
for all
sample
runs
Ext. masonry, 1671/
7" x 7", (VII), 1671/
22 ft. high 2
Forced air oil
furnace: 8.2%
usage
(average)
Family of three; husband works days; wife
runs daycare business at home with 2-6
additional children at home during weekday
working hours.
Home occupants use a mix of well seasoned
wood and relatively green wood, generally in
a one log wet to one log dry ratio.
Bypass rod of stove began indicating a
dark brown soot color in 86/87 season,
possibly indicating an inactive catalyst
according to the stove operation manual.
Home occupants report they were very
pleased with this catalytic stove's
performance.
-------
Table A-2 (Continued - Page 12)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #QF STORIES*3/ FREQ. OF USEC/ NOTES
V18
V31
L.E. (K)
for
sampling
runs 4-7
Cat. (P)
for all
sampling
runs
Prefab, metal,
6" dia., (I),
15 ft. high,
(samples 4-6);
18 ft. high,
(sample 7)
Prefab, metal,
6" dia., (XI),
15 ft. high
1040/
Approx.
3000/
2 +
basement
780/
15607
2
Zoned electric
baseboard
heat: 2.5%
usage average
in living room
zone.
Occasional use
in other
zones.
Zoned electric
baseboard
heat: 0% usage
in living
room.
Frequent use
in downstairs
zones.
Two adults live in home; this was the
project field office occupied by John
Sarsfield and Steve Mackey of OMNI.
Area heated by woodstove was basement
apartment of large three-level home.
Home occupants occasionally burned
considerable quantities of mill ends.
Home experienced smoke intrusion into home
and difficulty in starting new fires with
original 15 ft. high chimney. Start-up of a
cold stove usually required preheating the
stove pipe (by such means as irons and blow
dryers) and opening windows to overcome a
downward flow in the chimney.
Chimney was extended by three feet before
Sample 7 to virtually eliminate most smoke
intrusion problems. Windows were still
opened when starting new fires to help
draft.
Couple living in home; generally someone
home days.
Woodstove heats upstairs of home while
electricity serves heating needs of lower
level.
This stove was in use for one full heating
season prior to the start of the study in
fall 1985.
-------
Table A-2 (Continued - Page 13)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQQDSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIESb/ FREQ. OF USEC/ NOTES
V32
3=
I
Cat. (P)
for all
sampling
runs
Int. masonry,
7" x 11", (VI)
23 ft. high
V34 L.E. (M) Prefab, metal,
for 7" dia., (I)
sample 30 ft. high
runs 4-7
V35 L.E. (N) Prefab, metal,
for 6" dia., (I),
sample 23 ft. high
runs 4-7
N01 Cat. (A) Prefab, metal,
for all 8" dia., (I),
sampling 14 ft. high
runs
2240/
1120/
1 +
basement
1650/
1650/
2 +
basement
1350/
1350/
2 +
basement
1014/
1014/
1 +
basement
Zone electric
baseboard
heat: 0.0%
usage.
Gas forced air
furnace: 1.5%
usage.
Retired couple, generally home days.
Stove has hot water loop plumbed inside
firebox.
Inactive combustor replaced 10/86.
This stove was in use for one full heating
season prior to the start of the study in
fall 1985.
Retired couple, generally home days.
Occasionally burns considerable quantities
of mill ends.
Homeowner reported chimney fire 2/87.
Zoned electric Working couple; both spouses work days.
baseboard Home occupants report they were very
heat: 0.0% pleased with this low emission stove's
usage. performance.
Gas hot water
furnace: 0.0%
usage.
Retired couple, generally home days.
Flue condensation caused icicles at
chimney exit during cold weather.
Homeowner reports smoke intrusion into
home occasionally while operating stove in
catalytic mode.
Homeowner reports satisfaction with
catalytic stove, but is disappointed by
creosote build-up inside firebox walls and
loading door.
-------
Table A-2 (Continued - Page 14)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQODSTQVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
IQF STORIES5/ FREQ. OF USEC/ NOTES
N02
Cat. (D)
for all
sampling
runs
N03 Cat. (C)
for all
sampling
runs
Int. masonry,
7" x 11",
(VIII), 20 ft.
high
2208/
2208/
1 +
basement
Int. masonry,
7" x 7", (V),
24 ft. high
2000/
10007
1 +
basement
Oil forced air
furnace: 0.0%
usage.
Solar heating
system: served
as primary
heat source,
but was
difficult to
quantify.
Working couple; both spouses work days.
Stove was set up with a hot water loop on
the back surface of the stove.
Homeowner routed sheet metal ducting from
stove top surface to upstairs living area to
direct stove heat.
Firewood was exceptionally well seasoned
and dry.
Experienced catalyst substrate failure
3/86.
Field observations indicate this home
(especially basement with stove) was
frequently kept much warmer than average
during evening hours.
Single homeowner; runs business at home and
in separate shop.
Flue condensation resulted in condensate
leaking out of chimney clean-out door
occasionally.
Homeowner frequently allows home to remain
at below-normal indoor temperatures.
Woodstove used only as required when solar
heating could not meet heating demands.
Homeowner reports he was very pleased with
this catalytic stove.
-------
Table A-2 (Continued - Page 15)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIES*3/ FREQ. OF USEC/ NOTES
oo
N04 A/R (G)
for
sample
runs 1,
2, 3, 6,
7;
A/R (J)
for
sample
runs 4, 5
Ext. masonry, 2132/
7" x 7", (VII), 1422/
33 ft. high. 2 +
Approx. 5 ft. basement
of horizontal
flue pipe
before chimney
Oil hot water Family of four; both spouses work days.
furnace: 1.0% A sheet metal hood fits over the top of
usage. this stove and is ducted to a floor vent
upstairs.
Add-ons used on this stove were both
mounted in horizontal position because of
back exit flue configuration of the stove.
Homeowner reports smoke intrusion into
home while using add-on J and requested to
have the unit removed.
Home occupants report they were very
pleased with the performance of add-on G,
especially with regard to reduced creosote
accumulation.
Combustor found to be cracked, discovered
and replaced 2/86. It is not clear whether
the combustor was cracked when the unit was
installed or at a later date.
N05 A/R (F) Ext. masonry, 1722/
for 7" x 11", 1722/
sample (VIII), 2 +
runs 1-3; 31 ft. high basement
Conv. (0)
for runs
4-7
Gas forced air Family of four; one spouse working days
furnace: 85/86; both spouses working days 86/87.
frequent usage Homeowner reports using stove less during
reported. second heating season because of a change in
home occupancy patterns.
N06 A/R (I)
for
samples
1-3
Ext. masonry, 2496/
7" x 7", (VII), 12487
22 ft. high 1 +
basement
Gas forced air
furnace: 0.0%
usage.
Working couple; both spouses work days.
Home occupants moved from home between
study seasons and therefore dropped from
study.
-------
Table A-2 (Continued - Page 16)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE HOODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIES"' FREQ. OF USEC/ NOTES
N07
L.E. (K)
for all
sampling
runs
Ext. masonry,
8" dia.,
(XIII),
23 ft. high
1915/
11221
1.5 +
basement
I
I—»
IO
Gas forced air
furnace: 0.0%
usage.
Second
woodstove was
in use most of
heating season
during second
year of study
only.
Retired couple, generally home days.
Homeowner reports occasional smoke
intrusion into home. Occasional loss of
flue draft was also reported. Problem was
much less in second year.
Flue condensation resulted in condensate
leaking out of chimney clean-out door
regularly.
Home had a second woodstove upstairs while
the study low emission woodstove was in the
basement. During the 85/86 heating season,
this second woodstove was vented to the same
flue as the study stove, but was rarely ever
used. During the 86/87 heating season the
second woodstove was vented through a new
separate chimney and this second woodstove
was used as much as the study stove.
Home occupants report they were very
pleased with this low emission stove's
performance.
Approximately one quarter of the firewood
supply was cut in half to short 8-9 inch
lengths to fit the firebox of this low
emission stove.
Spray creosote cleaners were regularly
used on the stove flue/exit to prevent
creosote build-up leading to restricting
flow and smoke intrusion.
-------
Table A-2 (Continued - Page 17)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIES"/ FREQ. OF USEC/ NOTES
N08 Conv. (0) Ext. masonry, 1800/
for all 7" x 7", (VII), 1080/
sampling 25 ft. high 1.5 +
runs basement
Oi1 hot water
furnace: 0.0%
usage.
;E>
i
o
Retired couple, generally home days.
A sheet metal head fits over the top of
this stove and is ducted to a floor vent
upstairs.
Homeowner's burning practices as noted by
OMNI field personnel were to keep firebox
relatively full of wood and spin drafts
turned to near low. Homeowner also appeared
to continuously operate stove even in very
mild weather.
Chimney was swept approximately six times
per year due to rapid creosote build-up.
Chimney was equipped with a pulley system
chimney brush due to substantial creosote
build-up. This pulley/brush system was not
used during the study at request of OMNI
staff.
-------
Table A-2 (Continued - Page 18)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQODSTQVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY5/ #OF STORIES'3/ FREQ. OF USEC/ NOTES
N09 Cat. (B) Ext. masonry, 2024/
for all 7" x 7", (VII), 20247
sampling 19 ft. high 2
runs
I
ro
N10
Cat. (A)
for all
sampling
runs
Ext. masonry,
7" x 7",
(VIII),'20 ft.
high
30047
1502/
1+
basement
Gas forced air Family of four; both spouses work days.
furnace: 0.2%
average usage.
Zoned electric
baseboard
heat: 0.5%
usage average
in basement
zone. Daily
use reported
in other
zones.
Experienced flue condensation problems and
lower chimney ice formation while using this
catalytic stove. Creosote and condensate
leached into masonry blocks and ice would
form near the chimney clean-out door during
very cold weather. A light bulb was placed
inside in lower chimney section to help
reduce the problem.
Home occupants report they were very
pleased with the performance of this
catalytic woodstove.
Firewood used during second heating season
of study was dryer and this resulted in
reduced flue condensation problems.
Family of four; both spouses work days.
Flue condensation resulted in creosote
condensate leaching into masonry blocks of
chimney.
Glass in stove loading door broke and was
replaced 12/86.
Homeowner reports catalyst bypass would
bind when stove became very hot.
-------
Table A-2 (Continued - Page 19)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIESb/ FREQ. OF USEC' NOTES
Nil Cat. (D)
for all
sampling
runs
Int. masonry,
7" x 7", (V),
24 ft. high
1908/
1908/
2 +
basement
3=-
I
ro
ro
Zoned electric
baseboard
heat: 13.9%
usage in
kitchen.
Occasional use
reported in
other zones.
Retired couple, generally home days. Left
home for three-week vacation in February of
both heating seasons.
Homeowner reports bypass snags on stove
housing when stove is very hot.
Woodstove is located in small kitchen area
which can become very warm in efforts to
effectively heat the entire house. As a
result, the stove is often run at a lower
heat output to permit work in the kitchen,
and electricity is used in other rooms of
the home as the temperatures drop.
Homeowner reports dissatisfaction with
this catalytic stove, primarily because it
functions as a radiant heater when its
location in the home requires convective
stove heat.
Firewood is exceptionally well seasoned
and dry. Most of firewood was split into
smaller than average sized pieces.
-------
Table A-2 (Continued - Page 20)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
CO
GO
HOME
CODE WOODSTOVE
N12 A/R (I)
for
sample
runs 1-3;
A/R (J)
for
sample
runs 4-5;
Conv. (0)
for
sample
runs 6-7
N13 A/R (H)
for
sample
runs 1-3;
L.E. (M)
for
sample
runs 4-7
HEATED AREA/ ALT. HEATING
, FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIESb/ FREQ. OF USEC/
Prefab, metal, 1876/
8" dia., (I), 1876/
16 ft. high for 1 +
samples 1-4; basement
18 ft. high for
samples 5-7
Ext. masonry, 2080/
7" x 7", (VII), 2080/
22 ft. high 2
Oil hot water
furnace: 0.2%
usage average.
Zoned electric
baseboard
heat: 0.0%
usage. Use in
other zones
was reported
rare.
NOTES
Working couple; both spouses work days.
Experienced smoke intrusion into home
using add-on J usually in the form of
violent puff backs while in catalyst mode.
Homeowner reports occasional smoke
intrusion into home while using add-on I.
Homeowner reports it was necessary to
maintain "too hot of a fire for add-on to be
continuously activated."
Retired couple, generally home days.
Approximately half of firewood was cut
into small 8-9 inch length pieces to fit
inside low emission Stove M.
Glass in stove loading door broke and was
replaced 3/87.
-------
Table A-2 (Continued - Page 21)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE
N14
N15
WOODSTOVE
A/R (I)
for
sample
runs 1-3;
A/R (J)
for
sample
runs 4-5
Conv. (0)
for
sample
runs 6-7
L.E. (L)
for all
sample
runs
CHIMNEY3/
Ext. masonry,
7" x 7", (VII
22 ft. high.
Chimney
configuration
uses three 90
bends
Prefab, metal
6" dia., (II)
16 ft. high
HEATED AREA/
FLOOR SPACE/
#OF STORIES5/
2688/
), 2688/
2
O
, 1570/
, 15707
2 +
basement
ALT. HEATING
SYSTEM AND
FREQ. OF USEC/
Oil forced air
furnace: 0.0%
usage.
Gas forced air
furnace: 5.2%
usage.
Homeowner
reports daily
use of furnace
- approx. 1-2
ccf/day.
NOTES
Two to five home occupants, with someone
generally home days. Patterns of presence
by occupants were highly variable.
Experienced smoke intrusion into home and
soot plugging of combustor while using add-
on I .
Experienced smoke intrusion into home,
using Add-on J, in the form of violent
puffbacks while in catalytic mode.
Stove was frequently operated very hot and
near-cherry-red stove surfaces were observed
on several occasions.
Home was often kept at much warmer than
average indoor temperatures.
Firewood was generally larger than average
girth-size pieces.
Family of four; both spouses work days, but
often someone home during part of daytime.
Homeowner installed flue damper in stove
pipe after Sample 2 and reports use of the
damper for only the final portion of a burn
to preserve coals in stove.
Top section of metal chimney fell off on
several occasions and was entirely off for
Sample 7.
Large south-facing windows contribute
passive solar heating.
-------
Table A-2 (Continued - Page 22)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOQDSTOVE
HEATED AREA/ ALT. HEATING
. FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIES5/ FREQ. OF USEC/ NOTES
N16 Conv. (0)
for
sample
runs 1-3;
L.E. (N)
for
sample
runs 4-7
Ext. masonry,
7" x 11",
(VIII),
14 ft. high
11757
11757
1 +
basement
Zoned electric
baseboard heat
1.0% usage
average in
family room
zone.
Frequent use
reported in
other zones.
Family of four; one spouse works days.
Homeowner reports using less electricity
in second study year than the first heating
season of the study.
Home occupants report they were very
pleased with the performance of this low
emission stove.
>
ro
N18 Cat. (B) Ext. masonry, 1204/
for 7" x 7", (VII), 1204/
sample 21 ft. high 1 +
runs 4-7 basement
Zoned electric Family of four; one spouse works days.
baseboard Flue condensation caused some minor
heat: 0.1% creosote/condensate leaching into masonry
usage in blocks of chimney.
basement. Firewood was primarily large pieces of
Daily use white pine logs, generally 24 inches long
reported in and relatively large girth. Firewood was
other zones. exceptionally dry.
Field observations indicate homeowner
would regularly operate stove at high heat
outputs, sometimes resulting in warmer than
average indoor temperatures.
Home occupants report they were very
pleased with the performance of this
catalytic stove.
-------
Table A-2 (Continued - Page 23)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WOODSTOVE
CHIMNEY3/
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
#OF STORIES"/ FREQ. OF USE0/ NOTES
N32 Cat. (P)
for all
sample
runs
Prefab metal,
6" dia., (I),
21 ft. high
1440/
1440/
2 +
basement
Zoned electric
baseboard
heat: 0.0%
usage. Use in
other zones
was rare.
ro
en
Single homeowner working days; son stays at
house approximately one third of the time.
Stove was in use for one full heating
season prior to the study start in fall
1985.
This stove has an 8" diameter flue collar
and is being used with a tall, 6" diameter
chimney.
This stove is equipped with a three-inch-
thick combustor, while all other A stoves
had two-inch-thick catalysts.
Home occupants report they are very
pleased with the performance of their
catalytic stove.
Homeowner reports keeping stove thermostat
at same setting for the entire time he has
owned and operated the stove.
Homeowner would open doors and windows
regularly when house became too warm or for
additional ventilation.
House and stove were regularly kept much
warmer than average.
Homeowner reports operating stove
regularly so the supplied catalyst
thermometer read above 2000° F.
A significant portion of firewood used was
rotted wood, especially in the first study
heating season.
-------
Table A-2 (Continued - Page 24)
CHARACTERISTICS OF EMISSIONS-SAMPLED HOMES
HOME
CODE WQODSTOVE
HEATED AREA/ ALT. HEATING
FLOOR SPACE/ SYSTEM AND
CHIMNEY3/ #OF STORIES'3/ FREQ. OF USEC/ NOTES
N33
Cat. (P)
for all
sampling
runs
Int. masonry, 2194/
8" dia., (XII), 2194/
22 ft. high 2 +
basement
Gas forced air
furnace: 0.0%
usage.
Retired couple, generally home days.
This stove was in use for one full heating
season prior to the start of the study in
fall 1985.
Peeling of catalyst coating on combustor
noted 5/86.
Home occupants report they are very
pleased with performance of their catalytic
stove.
Most of firewood was poorly seasoned and
exceptionally wet.
Catalyst was replaced 3/87 after all
samples were collected. Homeowner reports
dramatically improved stove performance with
new combustor, suggesting the old combustor
used for both emissions tests had
experienced a lack of activity.
Notes:
a/See Table A-l for additional details of the chimney system.
^/Values in square feet. "Heated area" is defined as the floor space of the rooms or floors heated by
the stove. If the stove is located in an unfinished basement, with heat vented to upper floors, the
basement area is included as heated area but not included as floor space.
c/Alternative heating system is the method(s) of heating used in addition to the study woodstove heat
source. Frequency of use is defined as the percentage of time the alternative heating system in the
same room as the woodstove is in use. See Table 3-10A for more detail.
-------
Table A-3
SAFETY IMPROVEMENTS TO STUDY HOMES
In the first heating season of the study (1985/86), several fire code violations
and safety concerns were identified in study homes. In each case, the home
occupants were made aware of the problem(s) and encouraged to upgrade the safety
of their wood heating installation. Very few of the home occupants made any of
the recommended improvements, prompting study sponsors to make the safety
upgrades. These study-sponsored safety upgrades were performed in Fall of 1986
and included improvements to most of the major safety concerns in study homes
with identified deficiencies. Some safety concerns, such as two appliances
venting to the same flue, or chimney/lining damage, were considered beyond the
funding of this project. Safety improvements were not performed where the home
occupant found the recommended upgrade objectionable (i.e., floor or wall
protection not suited to home decor, repositioning of stove undesirable).
Following is a list of those study homes receiving study-sponsored safety
upgrades early in the 1986/87 heating season.
V03--A spark-retardant hearth rug was installed in front of the existing floor
protection to upgrade safety while using new front-loading stove.
V05--Shielding and stove pipe were installed to protect a cabinet near the stove
exhaust.
V10--The proper stove exhaust through the wall assembly was installed; walls,
ceiling, and curtains were shielded; and the proper chimney connection was
installed.
V17--A properly sized floor protection pad and wall shielding were added to this
installation.
V18--The chimney was extended to code height and a wind cap was installed to
reduce smoke intrusion into home.
V20--Shielding of a combustible wall and a chimney extension with cap were added
to this installation.
V21--Shielding of the ceiling and a larger floor protection pad were added to
this installation.
V26--A spark-retardant hearth rug was installed in front of the existing floor
pad and the chimney cleanout door was repaired to function properly.
V27--A spark-retardant hearth rug was installed in front of the existing floor
pad; chimney was extended with cap to code height; badly eroded stove pipe
and thimble were repaired.
V30--Shielding was installed to protect wood around thimble, wall, and curtains.
V34--Shielding of wall and a stove pipe transition piece were added to this
instal lation.
N05--Fireplace mantle heat shields were added to this installation.
N07--The thimble pipe was replaced with the proper unit in this home.
N11--A hearth shield and spark-retardant hearth rug were added to this
instal lation.
N12--The chimney was extended to code height.
N13--A spark-retardant hearth rug was installed in front of the existing floor
protection at this installation.
N17--A heat shield and adequate floor protection were installed at this home.
N18--The chimney thimble was replaced with the proper unit.
N20--A close clearance thimble (through the wall) and heat shields were installed
at this home.
A-28
-------
For further information
on NYSERDA reports or
publications contact:
Department of Communications
NYS Energy Research and
Development Authority
Two Rockefeller Plaza
Albany, N.Y. 12223
(518) 465-6251
-------
Energy Authority Report 87-26, Vol. I
State of New York Mario ML Cuomo, Governor
New York State Energy Research and Development Authority
William D. Cotter, Chairman Irvin L. White, President
------- |