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

-------