EFFECTS OF DESIGN  FACTORS  ON  EMISSIONS FROM NON-CATALYTIC

                  RESIDENTIAL  WOOD COMBUSTION APPLIANCES
PREPARED BY:
OMNI ENVIRONMENTAL SERVICES, INC. UNDER SUBCONTRACT
WITH RADIAN CORPORATION
PREPARED FOR:   EMISSIONS STANDARDS AND ENGINEERING DIVISION
               EPA CONTRACT NO.   68-02-3816
                  U. S. ENVIRONMENTAL PROTECTION AGENCY

               OFFICE OF AIR QUALITY PLANNING AND STANDARDS

              RESEARCH TRIANGLE PARK, NORTH CAROLINA  .27711
                                 MAY 1986

-------
                                  DISCLAIMER

This report has been reviewed by the Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency, and approved for publication
as received from the Radian Corporation.  Approval does not signify that the
contents necessarily reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names of commerical products
constitute endorsement or recommendation for use.

-------
                                      WOODSTOVE DESIGN FACTORS







 SUMMARY





     The  combustion  of wood in naturally drafted  non-catalytic



 Residential Wood Combustion (RWC) stoves involves highly  complex



 chemical  processes  which  are sensitive to a  wide  variety  of



 influences.  Key  elements required for efficient combustion  in-



 clude  high combustion zone temperatures,  adequate air (oxygen)/



 good  air and fuel mixing,  adequate residence time,  and  appro-



 priate air/fuel ratios.   The batch process of wood combustion in



 the naturally drafted cordwood stove presents special problems in



 that  the entire fuel charge is involved in various and  changing



 stages  of a complex combustion process throughout the fuel  load



 burning cycle.   Ideal conditions vary during each stage,  making



 complete  and efficient combustion of the entire fuel charge in a



 single  stove configuration very  difficult.



     A  variety of existing stove technologies are  examined  for



 their effectiveness in emissions reduction.   Theoretical  consi-



derations  are followed by review of supporting data and observa-



tions.   The  existing data base is sparse,  and positive conclu-



sions  cannot  be  made.     However,  the  following  trends  are



observed:



     1.   Emissions  are lower  from stoves with smaller  fireboxes



         than similar stove designs with larger fireboxes.



     2.   Low  firebox  heights  appear  to  contribute  to  lower



         emissions.



     3.   Air  inlets  above the fuel load in the  firebox promote



         good air fuel mixing  and reduce potential emissions.

-------
                                 WOODSTOVE DESIGN FACTORS
4.  Air entering the firebox near or up through the coal bed
    ("underfire air") results in higher emissions.
5.  Exhaust   exits  located low in the  firebox  appear  to
    cause high emissions.
6.  Preheated secondary air/  or more properly termed, addi-
    tional  combustion air,  introduced at high  temperature
    locations appears to reduce emissions.
7.  Thermostatic air supply controls on non-catalytic  wood-
    stoves  often  cause air-starved conditions at high fuel
    load  and firebox temperatures as the thermostat  cycles
    closed.    These conditions  can significantly  increase
    emissions.
8.  Pellet-fueled  stoves  utilizing  mechanically  assisted
    drafts  have demonstrated emission rates below the  most
    efficient  catalytic stoves.  Densified fuel in  natural
    draft  cordwood stoves also exhibit significant emission
    reductions.

-------
                                       WOODSTOVE  DESIGN FACTORS

                         TABLE OF CONTENTS
 I.    Introduction                                        1
 II.   Woodstove  Combustion  Fundamentals                  4
 III.  Woodstove  Design  -  Natural  Draft  Cordwood           9
      A.  Baseline Stoves                                 10
      B.   Firebox Geometry                               11
           1.  Firebox  Size                             11
           2.  Firebox  Shape                             19
           3. Pathway of  Combustion  Products             21
                a. Baffling                             24
                b. Combustion  Chambers
                     - Primary & Secondary              26
                c. Downdraft Combustion                 27
      C.  Combustion Air - Distribution,
                Conditioning/  and Control               29
           1. Primary Air                               29
           2. Onderfire Air                             30
           3. Secondary Air                             32
           4. High Minimum  Burn Rates                    33
           5. Controls                                   34
     D.  Heat Exchange  Systems                           36
     E.  Construction Materials                         39
IV. Other  Non-Catalytic  Technologies                    43
     A.  Pellet Fuel                                     43
     B.  Densified Fuel                                  46
     C.  Mechanically Drafted Cordwood  Stoves            46
     D.  Microprocessor Control                          47
References
Appendix A - Stove Descriptions
Appendix B - Stove Test  Data
Appendix C - Graphical Data Presentation

-------
                                      WOODSTOVE DESIGN FACTORS

 I.  INTRODUCTION
 The  United States has experienced a resurgence  in the popularity
 of  the woodstove or residential wood combustion  (RWC)  appliance
 over the past 15 years.   This growth in the use  of wood fuel  to
 meet  home heating demands is shown in Figure 1  and represents an
 annual consumption  of  approximately 45  million cords   (1, 2).
 While  wood  still ranks a distant third as  a   household  energy
 source  (Table 1),  the volume and nature of woodstove  emissions
 make it a source of significant environmental concern.

                           TABLE  1 (3)
                        ENERGY SOURCE FOR
                          HOME HEATING
                            (percent)
                 Natural Gas	49.5
                 Electricty	24.8
                 Wood	9.1
                 Oil	7.0
                 Other	9.9

     A  1984  study by Arthur D.  Little for  the U.S.  Consumer
Products  Safety  Commission  shows that there are more  than  27
million wood heating devices in U.S. homes (Figure 2).  Of these/
approximately  13 million are freestanding and   fireplace  insert
woodstove  appliances.   Most estimates of wood  use indicate that
the  use  of this fuel in U.S.  homes will continue  to  increase
through this century.   While the sale of woodstoves has  leveled
off  at  approximately  1.2 million units per  year   (Figure  3),
reports by "Wood n' Energy",  an industry trade  journal, indicate
that 63% of the purchases are for first stoves,  14% are second or

                                                       Page - 1

-------
                            Figure 1




      NATIONAL ESTIMATES OF WOOD FUEL USE (1860 - 1980)  (1)
3.0 -,
   1860
1880
1900
1920



Year
1940
1960
1980

-------
                           Figure 2
                  WOOD HEATERS IN USE (A)
                         1984  Data
s
«
v
WOOD HEATERS IN USE

15 Million
FIREPLACES

75 Million
STOVES

4 Million
INSERTS

r 1.2 Million
FREESTANDING
FIREPt ACES/STOVES

                          Figure 3
          SALES OF RESIDENTIAL WOOD STOVES
  1200

  2000-
  U*001
  MOO-1
wuoo-
o
I
  «oo-H
  400-

  JOO-
Estimated Wood Stove
   Sales & Imports
     Tl
         73  13  14  7J
                            77  "7«  7»  "$0
                                                •62   -03  «4

-------
                                       WOODSTOVE  DESIGN  FACTORS

 more purchases/ and 23% represent replacement  stoves  (4).
      Of  more importance as an environmental concern  are the  air
 pollutant  emissions  generated by this   source   category.  Wood-
 stoves  produce large amounts of particulates  (smoke) and   carbon
 monoxide  at  levels that present serious concerns  in many  urban
 areas.   Recent  estimates indicate that  woodstoves are currently
 producing  approximately  2.5 million  tons of  particulates   (5).
 These  emissions  constitute a major contributing  cause  of  air
 quality violations and interfere with  the achievement and mainte-
 nance of national ambient air quality  standards  in  several  areas.
 These conditions have been well documented in  Missoula, Montana;
 Portland  and  Medford,  Oregon;  Reno,   Nevada;    Boise,   Idaho;
 Spokane/   Washington?   Juneau,  Alaska;    Denver,  Colorado; and
 Albuquerque/   New Mexico.  High air quality impacts from RWC  have
 been  identified in the urban air of most  eastern,  northern/  and
 western states.
      The   physical  and  chemical  characteristics  of  woodstove
 emissions  further accentuate this source  problem.  The particu-
 lates   are predominantly submicron organic condensate  materials.
 Several of the  organic compounds have demonstrated  carcinogenic
 and mutagenic properties.    A recent study showed stove emissions
 to  be  highly acidic (pH 2.5 to 3.8),  with high persistance  in
 acidity due to organic acids (6).
     These documented conditions and  urban problems  have   caused
 three  states  (Oregon,   Colorado/  and  Montana) to adopt specific
 stove   emissions  regulations.    Several  other states are in  the
process  of  developing  strategies to address   these  problems.
This   report  provides  a detailed discussion of the  technology  of
                                                        Page - 2

-------
                                      WOODSTOVE  DESIGN  FACTORS



 non-catalyst,   natural draft woodstoves.  The theory of woodstove

 combustion,  the   chemistry and physics of  combustion,  and  the

 state   of  RWC  woodstove design for  improved efficiency and  low

 emissions are addressed.

     The  report presents a synopsis of woodstove combustion fun-

 damentals  and  discussions on the effects of  combustion  chamber

 design  on  the emissions and efficiency characteristics of  non-
                                         0
 catalytic  woodstoves.   All reported  values,   unless  otherwise

 noted,  were  obtained at the laboratories of OMNI  Environmental

 Services,  Inc.,   using  the  Oregon Department  of  Environmental

 Quality   "Standard  .Method  for  Measuring  the Emissions   and

 Efficiences of  Residential Woodstoves", June 8,  1984 edition.  It

 is important to note that other fuel loading densities  and confi-

 gurations,  efficiency algorithms, particulate measuring systems,

 laboratory altitude,  or stove operating practices will result in

potentially  significant differences in reported efficiency  and

emission values.  Care should be taken when comparing data .gener-

ated using different procedures or methods unless equivalency has

been demonstrated and all correction coefficients are applied.
                                                       Page - 3

-------
                                      WOODSTOVE DESIGN FACTORS







 II.  WOODSTOVE COMBUSTION FUNDAMENTALS










     The  burning of wood fuel under conditions found  in the fire-



 boxes  of  RWC woodstoves  involves a complex  combustion  process



 which pyrolyzes and oxidizes volatile,  semi-volatile,  and solid



 components of the wood material.   The engineering rule of  thumb



 is   that  the four conditions or elements of combustion  must  be



 optimized  for this combustion process to take place  efficiently:



 time, temperature, turbulence, and the air/fuel mixture ratio.



     Time is required to allow thorough air and fuel  mixing,  for



 energy-releasing  chemical reactions,  and for heat   transfer  to



 occur.   If  the  residence time of the fuel and oxygen  at  high



 temperatures is too short, combustion will be incomplete and heat



 transfer  from the gases to the medium being heated will be inef-



 ficient.   If residence time is too long,  gas velocities in  the



 combustion  chamber  will be low,  and the driving mechanism  for



 mixing of the air and fuel gases will be weak.   This also  leads



 to incomplete and inefficient combustion.



     Temperature  is  important since the rates of  the  chemical



 reactions,   which  are  the  essence of the  combustion  process,



 increase exponentially with temperature.  Generally,  high combus-



tion temperatures ensure complete combustion.  However, in indus-



trial gas,   oil,  and coal-fired systems, and in internal combus-



tion engines,  excessively high combustion temperatures can  also



cause  nitrogen oxides pollution.   Conventional residential wood



combustion    typically   occurs  at  relatively  low   combustion



temperatures  because overall-air/fuel ratios are generally  very





                                                       Page - 4

-------
                                       WOODSTOVE DESIGN FACTORS

 air-rich/  and because the combustion chamber  also  serves  as  the
 heat exchanger.
      There  is  an  important "feedback"   mechanism  at work   in
 combustion chambers.   Hot gases which are burning,   or have been
 nearly  completely  burned,  must mix with and transfer heat   to
 fresh  fuel and air so that the combustion process  is  sustained.
 Other  ways  to  accomplish this internal  heat  transfer  are   by
 radiation  from  refractory combustion chamber walls  to the  fuel
 and  air,   and by contact with hot coals.    When the heat transfer
 rate  to  the fresh mixtures is enough to maintain   ignition  tem-
 peratures,  a "stabilized" flame is produced.
     Turbulence is an important factor because  the fuel  gases  and
 air  must  mix, and as just indicated,  the burning and  burnt mater-
 ials  must  mix with the fresh fuel and air.    Of the  two types  of
 mixing,   the first type of mixing is  termed "air/fuel mixing" and
 the   second  type is termed "age mixing" because of the old  and
 young cbaractistics of the mixing constituents.   With a close
 look   at  either  of these types of mixing,  it is  observed  that
 different special scales are involved.  The swirling  and fluctua-
 ting   motions seen in a gaseous combustion  flame illustrate this.
 These turbulent  gases contain different size  scales,   from  vor-
 tices  almost as  big as the whole combustion chamber,   to  eddies
 down   to  millimeter (mm)  size.    As  the size  scale of   turbulent
 eddies  decreases,   the gases are mixed more and more intimately.
To have chemical  reaction of the gaseous mixtures,  oxygen mole-
 cules  must collide with fuel molecules and hot gas or  one of the
other  ignition  sources must be available to provide the heat  and

                                                        Page - 5

-------
                                      WOODSTOVE  DESIGN  FACTORS

 reactive  species  for  ignition.   The turbulence  is very important
 because  it  promotes molecular contact.    Nearly all   industrial
 combustion chambers/  and many residential  combustors are  "mixing
 controlled."   That is,  the rate at which  the   mixing   processes
 occur which bring the fuel gases,  air,  and  hot gases  into  inti-
 mate  contact,  controls the overall rate of  burning:    "if   it's
 mixed,  it's burned."
     Mixing   can   be  produced  by   mechanical   blowers,   by
 piston/valve action in reciprocating engines, by flow pulsations,
 and  by natural draft.   Natural draft occurs because   a  slight
 suction or negative pressure is produced in  a firebox  due to the
 lower   density (buoyancy) of the hot combustion  gases.   Natural
 draft   is  the  weakest driving force  of   the   available  mixing
 schemes,   and   provides  the  most  difficult  process  control
 challenges.  The current challenge in wood  combustor design  is to
 use the small amount of available natural draft  as effectively as
 possible,  or  to incorporate a mechanical  draft system which  is
 reliable  and  compatible  with wood  combustion aesthetics  and
 objectives.
     Air/fuel  ratio  is important to  efficient combustion  for
 several  reasons.   Generally  the overall  air/fuel ratio  is  in-
 creased  above the chemically correct (stoichiometric)  value  for
 fuels which are more difficult to burn.   For industrial  combus-
 tors, the excess air required to burn natural gas is about 5%, to
burn oil about 10 to 15%, and to burn pulverized coal about  20 to
 25%.   That is, just enough excess air is added  to make sure  that
all  of  the fuel molecules find oxygen molecules with   which  to
 react.   Usually this ideal amount of excess air  can be  identified

                                                        Page  - 6

-------
                                       WOODSTOVE DESIGN FACTORS

 by looking for the "knee" in the curve of carbon monoxide exhaust
 emissions versus air/fuel ratio.  For industrial wood burners the
 recommended  amount  of excess air is not quite as  well  defined/
 though  well engineered systems appear to operate best on  50  to
 100%  excess  air.
      Too much excess  air is detrimental to efficiency/   since  as
 noted  above,  excessive air leads to low flame temperatures  and
 inefficient  oxidation  of the fuel;    that  is, the  combustion
 efficiency  is affected.   Combustion efficiency is a measure  of
 the   completeness of  the combustion process or  the  conversion  of
 the   chemical energy  of the fuel to the heat of the burned gases
 in   the  firebox.
      Too  much  excess air also affects  thermal efficiency,  or
 overall efficiency,   which is the percentage of the fuel chemical
 energy  which is actually transferred to the medium being heated.
 If   the excess air amount is too high,   the increased  flow rate
 carries  a higher proportion of the liberated chemical  energy  up
 the   exhaust   stack,   thereby decreasing the thermal  efficiency.
 Furthermore,   less time is available  for heat transfer  because of
 the   reduced  residence time,  and the heat transfer potential  is
 reduced because the temperature of the  burnt gases  is diluted  by
 the  excess air.   Thus,  the   rule of  thumb is to use only as much
excess   air  as is necessary,   unless excess air must be used  to
prevent overheating of the chamber materials.
     Although  all industrial  and residential combustors  operate
overall   air-rich,  local  pockets of  fuel-richness  unavoidably
exist.    These   pockets  are identified with carbon monoxide  and

                                                        Page -' 7

-------
                                      WOODSTOVE DESIGN FACTORS





 aerosol   formation.    Flames in these regions usually exhibit  a



 yellow color due to thermal radiation from soot aerosol.   If the



 flame is  well mixed,  and thus well aerated throughout,  its gases



 will appear blue in color.   Sometimes this blue color is  diffi-



 cult  to  see,  however,  because it is overcome by  red-to-orange



 thermal radiation from the hot soot aerosol.



     Even though conventional residential wood combustors operate



 in  most  cases    very air-rich,  much of the excess  air  never



 becomes intimately mixed with the fuel gases,  and thus,  because



 of  poor mixing,  the flame is yellow in  appearance,  containing



 soot  and unburned tar aerosols.     Even if the excess  air  does



 become mixed with the other gases, the resulting air-rich mixture



 is frequently too cool to react completely.



     In order for fuel to burn in a woodstove at all,  all  wood-



 stove  designs to some extent have to provide  conditions for the



 four elements of combustion discussed here.   The extent to which



 each of these elements is optimized dictates the combustion effi-



 ciency  and emission characteristics of a stove.   The   following



 sections describe various design factors which have been found to



 affect  the combustion process,  and involve  some or all of  the



basic  combustion  elements in naturally  drafted,  non-catalytic



woodstoves.
                                                       Page -  8

-------
                                      WOODSTOVE DESIGN FACTORS

 III.  WOODSTOVE DESIGN - NATURAL DRAFT CORDWOOD

      "Traditional"  or first/second generation woodstove  designs
 represent  toe  vast majority of stoves in use and a  substantial
 portion  of  stoves currently on the market.   This  category  of
 stove  is "naturally  drafted",  in that combustion air is  drawn
 into the firebox by the low pressure created in the stove by  the
 column  of  heated buoyant gases rising in the chimney.   A  vast
 array of styles/  shapes/  configurations/ gas flow patterns/ air
 distribution systems/  construction materials and sizes of stoves
 are available.  Manufacturer  claims of high efficiency or desir-
 able burning characteristics are common/  while little documenta-
 tion beyond personal testimony is often available.   This section
 assesses  these design considerations relative to the effects  of
 stove design on combustion efficiency and pollutant emissions.
     Documentation  used  in this report is presented  under  two
categories:
     1)    Data  collected  through the certification test  proce-
          dures  required  by the Oregon DEQ Standard Method  for
          Measuring  the Emissions and Efficiency  of  Woodstpves
          (June 8, 1984).  These data represent averages measured
          over  entire burn cycle tests (i.e./  the burning of an
          entire fuel load);  and/
     2)    Data collected through non-certification testing. These
          data  generally  represent  discreet burn  segments  or
          partial burn cycle results. References are measurements
          made   with   calibrated  instrumentation  or   reflect

                                                          PAGE 9

-------
                                       WOODSTOVE DESIGN FACTORS

          qualified   information  collected  by  observation    of
          phenomena during controlled test conditions.
      Descriptions and specifications of all stoves  referenced   in
 the  text are provided in Appendix A.   Appendix A  also lists the
 alpha  character stove code (A through FF)  which is used  in  the
 text  to reference each model tested.   Stove data  are referenced
 by the stove code and run number (e.g.,  Bl-1).   Appendix B con-
 tains a table of certification test data listed by  stove code and
 run numbers.   Appendix C contains graphical presentation of  se-
 lected stove test data.
         Where  comments in the following discussions are based  on
 partial test cycles or observations in the  laboratory,  reference
 is  made to  this effect.

 A.   BASELINE STOVES
      The relative  performance of newer technology stoves  pre-
 sented   in  this report is based on comparisons  to conventional  or
 baseline type appliances.   While  a wide variety of conventional
 box-type stove  models are in use and currently on the  market,
 many  are  of a similar size and technology.    These stoves  typ-
 ically   have  a 2.0 to 3.5 cubic foot firebox,   have  manual  air
 controls located  on  firebox sides or doors,   and   have  limited
 internal baffling.    Construction materials  are typically plate
 steel   or  cast  iron  with or without fire-brick lining  in  the
 firebox  (Figure 4).
     Laboratory testing was conducted on conventional   woodstoves
with  medium and  large fireboxes to provide baseline data for the
development   of woodstove regulations.    These  data are  used   to

                                                           PAGE  10

-------
                                                Baffle
Air Inlet
                      FIGURE    4




             BASELINE (BOX TYPE) WOQDSTOVE

-------
                                       WOODSTOVE DESIGN FACTORS

 define  the  baseline  performance capabilities  of   conventional
 stoves for comparison to newer technology stoves.  These  compari-
 sons  provide  relative measurement of  the  emission  reductions
 obtained with the newer woodstove technology.    Stoves Z, AA, CC,
 DDf   and EE are typical of baseline stove technology.  Emissions
 from  stoves in this group ranged from 9 to 48  grams   particulate
 per  hour.

 B.   FIREBOX GEOMETRY
      Firebox geometry describes the size and shape of the firebox
 and  the pathway of combustion gases in a stove  from the air  inlet
 to the flue collar exit.   Firebox geometry represents one of the
 largest  variables in woodstove design and also represents one of
 the   most  significant factors in the  resulting efficiency and
 emissions characteristics of a stove.  Firebox  size refers to the
 measure  of a stove's volume,  while firebox shape describes the
 relationship  of wall,  door,  floor,  and top   dimensions.   The
 pathway of  combustion products in the firebox,   through the  stove
 and   up the flue,   is a function of firebox geometry,   air   inlet
 location and baffling. .

 1.    Firebox  Size -  The capacity of a woodstove to  hold fuel is
 determined   by the size,  or usable volume, of  the firebox.  Many
 manufacturers report that the U.S.  market requires large  stoves
 which  enable consumers to load their stoves with large amounts of
 wood.  The  ability to maintain long burn duration without reload-
 ing  or   adjusting air settings is commonly regarded  as a   major
marketing   feature.    As a result,  most stoves sold  in the  U.S.
market have  firebox  volumes of about 2 to 6 cubic feet (57 to 170

                                                          PAGE 11

-------
                                       WOODSTOVE  DESIGN  FACTORS

 liters)   based  on  test  data  from OMNI  Environmental and observa-
 tions  at  national  trade shows.  Several manufacturers  from  Europe,
 where   smaller   stoves  used  as  room heaters  are   the  rule,  have
 reported   increasing  the size  of stove models intended for  the
 U.S.   market.   The capacity  for an  "overnight burn"  is perceived
 by many manufacturers as a key  design element.

     Firebox  size can also limit the  maximum potential  heat
 output from a  stove.   If more wood can be  placed in a stove and
 if air supplies  are sized accordingly  larger,  a greater mass  of
 wood can.be burned in a given amount of time.    Larger  stoves, in
                            «
 addition   to having greater  fuel capacities,  have greater  total
 surface area for greater heat exchange potential.    A  high heat-
 ing capacity has been cited  by  some manufacturers as being neces-
 sary to deliver enough  heat  to  older,  poorly insulated houses in
 colder climates.
     Another variable,  often not considered by  manufacturers and
 consumers, is the  true  space heating capability  of a stove.  Most
 stoves  are marketed  and sold  on the  ability  to heat  an  entire
 house.  The installation of a stove in a single  room, with little
 or  no forced or convective air circulation,   often reduces  the
 heating  effectiveness  of an appliance to only  a portion of  the
 total  house.   Practical limits of heating  an entire house  with
one stove do exist due to maximum "comfort range" temperatures in
the  room  with the stove.   Adjacent  or distant rooms  in  these
cases  are usually significantly cooler.   This  can be  important
in terms of pollutant emissions,  as stoves  capable of  supplying
total heat requirements for an  entire  house  would be at lower and

                                                          PAGE 12

-------
                                       WOODSTOVE DESIGN FACTORS

 less  efficient burn rates to maintain comfort range temperatures
 in the room with the stove.
      The effect of firebox and fuel load size on particulate  and
 carbon monoxide emissions was investigated by Burnet and  Tiegs  in
 1983.    A  model was being sought with which to predict emissions
 without conducting extensive actual laboratory testing.  A number
 of performance parameters were investigated in efforts to find  a
 single prediction factor.
      Data  from  existing stove tests,  all conducted  using  the
 Oregon  test procedure,   were reviewed.   Several stove types with
 baffled  and  non-baffled   fireboxes,  and with and without sec-
 ondary  air  and secondary combustion  chamber designs  were  com-
 pared.    Five stoves,  each with a different firebox volume, were
 selected,   representing  a variety of stove types (Stoves  Z,  AA,
 BB,  CC,  DD,   and E). Firebox sizes ranged from 0.8 to 3.5 cubic
 feet  (23  to 100 liters),  representing  stove sizes from very small
 to  relatively large.
     The  effect of burn  rate on particulate emissions from  each
 of the  five stoves is readily apparent in  Figure 5.    The smaller
 firebox  sizes  show consistently lower emissions across the range
of  heat  outputs.   When combustion efficiences from the  five
stoves  were examined relative to burn rate,   a similar relation-
ship  was   observed (Figure 6).    The  larger stoves   show poorer
combustion  efficiency,   which indicates higher release rates   of
carbon monoxide  and unburned organic species across  the range   of
burn rates.
     With   small   firebox stoves demonstrating higher   combustion

                                                          PAGE  13

-------
  60
  50
o>
o»
E
o
2
O
H 30
ui
UJ

!5 20
o
H
cc
Q. 10
 FIREBOX
 (ft3)
 a 0.8
 A 1.4
*2.0
 • 2.4
 • 3.5
STOVE
CODE
  BB
  CC
  DD
   Z
  AA
                               6        8       10       12        14
                                BURN RATE (Ibs/hour)
                                      Figure 5
                        The effects of burn rate on particulate
                        emission factors for RWC stoves with various
                        firebox sizes.
       16

-------
65
                            3        4        5        6

                               BURN RATE  (Ibs/hr)

                                    Figure 6

                      The effects of burn rate on combustion
                      efficiency for RWC stoves with various
                      firebox sizes.
8

-------
                                       WOODSTOVE  DESIGN FACTORS

 efficiences   and  lower  emissions,   the factors controlling   these
 parameters were  investigated.   Firebox temperatures  were  believed
 to be  a critical  factor,   with  higher  temperatures yielding   high
 combustion   efficiences.   Standardized  firebox  temperature   data
 were   not available.    Flue  temperature was  therefore  used as  an
 indicator of firebox  temperature.   While  the heat transfer  effi-
 ciency of the stove will affect flue temperatures,   an indication
 of firebox  temperature is still possible.   Figure  7  shows   that
 the  smaller stoves  (0.8 and  1.4  cubic  foot)  demonstrate in-
 creasing  combustion  efficiency with decreasing  size  and in-
 creasing  burn  rates as indicated  by  flue temperatures.   The
 larger  stoves,   however,  are  less sensitive to firebox  size ef-
 fects,  and   their results for  combustion  efficiency versus   burn
 rate lie on  an almost universal curve.
     Particulate  emissions were compared  with flue  temperatures
 (Figure 8).   As  the combustion efficiences  indicate,  lower  emis-
 sions  were  exhibited by the smaller stoves  across the range  of
 flue temperatures.   The larger stoves  showed a  relatively  close
 grouping.    These data  indicate that  both combustion  efficiences
 and  particulate  emissions were improved with smaller stoves  at
 any given flue temperature.   All stoves showed  improved  perfor-
mance at higher flue temperatures and  higher burn rates.
     From  the data noted above,  it was apparent that burn   rate
 and  combustion  temperatures were  strong factors in  determining
particulate emissions from a stove  of  any size.   The   importance
of firebox size appears to be a  function of  two  parameters:  the
proportion  of  the  combustion  chamber that  is  maintained  at
elevated temperatures at any given  time  and  the size of the   fuel

                                                          PAGE  14

-------
95
90
85
UJ

iZ
u.
UJ

O
K

CD
S
O
o
80
75
   70
 65
                                       o—
                                             I
                                  _L
                                                                I
                                                                FIREBOX
                                                                 (ft3)
                                                                a 0.8
                                                                1 1.4
                                                               [*2.0
                                                                • 2.4
                                                                • 3.5
                                                      STOVE
                                                       CODE
                                                        BB
                                                        CC
                                                        DD
                                                         Z
                                                        AA
           IOO
200     300      400     500     600
      FLUE  GAS  TEMPERATURES   (°F)
                                                               700
                                     Figure  7
               The effect of firebox temperature  (as indicated by flue
               temperature) on combustion  efficiency in RWC stoves with
               various firebox sizes.

-------
  50
  40
E
o
  30
CO
CO
CO
2 20
UJ
   10
                                   FIREBOX
                                    (ft3)
                                   a 0.8
                                   A 1.4
                                  *2.0
                                   • 2.4
                                   • 3.5
                                    0.8
  STOVE
   CODE
    BB
    CC
    DD
     Z
    AA
             IOO
200     300      400      500      600
AVERAGE  FLUE TEMPERATURE  (°F)
700
                                  Figure 8
              The effects of firebox  temperature (as indicated by flue
              temperature) on emission factors in RWC stoves with various
              sizes.

-------
                                       WOODSTOVE DESIGN FACTORS

 load.   A smaller stove will maintain flame temperatures  through-
 out  a larger fraction of the firebox than will a large stove   at
 the same heat output rate.  As the heat output rate is increased,
 the  proportion  of the firebox at "flame temperature" increases
 for   all  stoves,   but will always be higher for  small  stoves
 until  some  point is reached for each stove at which the  entire
 firebox is at flame temperature.
      For equivalent burn rates,  mixing of gases and air   appears
 to  be more efficient in a small firebox. This is indicated by the
 lower  emissions  and  higher combustion  efficiencies measured.
 Flow   velocities and turbulence would be greater in these  cases
 due  to the higher temperatures and smaller  spatial  dimensions.
 Smaller   fireboxes  also  theoretically  allow  more  efficient
 air/fuel  mixing,  due  to  smaller combustion  chamber  volumes.
 Smaller dimensions reduce the distance combustion air must travel
 to  reach all areas in the firebox.
      The size of the fuel charge appears to be the primary criti-
 cal factor,   with smaller loads generating lower emissions in the
 same  stove at the same burn rate.    The batch process involved  in
 fueling  a stove requires an entire fuel charge to be  placed   in
 the  firebox  at  once.   As the fuel  load is heated,  gasification  of
 the  wood occurs.    The larger the fuel load,   the more  wood   is
 subjected  to gasification,  resulting in greater  quantities   of
 fuel  gas being  released over a  given time.    At a  fixed heat
output   level,   more  fuel gas will  be  released from a large fuel
charge   than  from a small charge.    Lower mixing intensities and
more cool  spots  in  larger stoves will  result  in higher emissions,

                                                           PAGE  15

-------
                                      WOODSTOVE DESIGN FACTORS
especially under low fire conditions.  This effect is illustrated
in  Figures  C-l through C-5 in Appendix C.   These  graphs  plot
emission  rates from various woodstoves with and without  firebox
size modification as a function of heat output.
       To  provide  a basis for predicting  emissons  from  stove
design  criteria,  a  method  of normalizing data from  all  non-
catalytic stoves was still required.   In light of the  available
information,  a measure of relative burn intensity or an index of
the vigorousness of a fire in each stove type was developed.  The
index  is based on the percent of the fuel load burned per  hour.
This  approach permits direct comparison of different stove sizes
as  a  function of their relative fuel  loading  capacities.   It
should be noted that since the Oregon test procedure calls for  a
fuel loading density of 7 pounds of fuel per usable cubic foot of
firebox  volume,  each stove is loaded according to its  specific
size for certification testing.
    The percent of fuel load burned per hour provides an  indica-
tion of the fraction of the fuel load that is involved in combus-
tion at a given time.  An example is presented in Table 2.
                                                          PAGE 16

-------
                                      WOODSTOVE DESIGN  FACTORS
       TABLE 2:  PERCENT  OF FUEL LOAD BORNEO FOR DIFFERENT

          FIREBOX SIZES OPERATING AT THE SAME BORN RATE
                         STOVE  1     STOVE 2     STOVE 3

Firebox Volume
(Cubic Feet)                123

Fuel Load
(@ 7 Ib/Cubic Ft.)       7  Ibs.     14 Ibs.     21 Ibs.

Burn Rate (7 Ib./hr.)       7           7           7

Percent Fuel Load
Burned/Hour              100          50          33
                                                          PAGE  17

-------
                                       WOODSTO.VE  DESIGN FACTORS

      If each stove  in  Table  2  burns  at 7  Ibs.  per  hour,   at  the
 end  of  one hour the  2  cubic  foot stove  (Stove  2)  burns  half   of
 its fuel load,   and the  3  cubic  foot stove  (Stove 3)  burning  one
 third  of its fuel  load,   while the 1 cubic foot stove (Stove  1)
 has consumed the entire  fuel charge.
      Emission Factors  (gram/kg)  for  each  of the  five  firebox size
 stoves   are   plotted  as a function  of the percent  of  fuel  load
 burned   per  hour (Figure 9).   The curve shows that  emissions  in-
 crease   dramatically in each stove regardless of firebox  size   at
 burn  rates less  than about 40% of fuel load burned  per  hour.
      A  significant  aspect  of these results is that  the  data  base
 for  determining the effects of  firebox size on  particulate emis-
 sions   was   19 test  runs on  5  different stove  types   and sizes.
 Test  data are now available  on a larger number of stoves  although
 all   recently tested  stoves  are within  the  mid-  to  lower-size
 range   of the original test  group (i.e.,  2.5 to 1.2  cubic feet).
 All non-catalytic stoves undergoing  certification testing at  the
 OMNI lab through August,   1985,  have  had  a firebox  volume of less
 than 2.5 cubic feet.
     The  original 19 data points represented designs typical   of
 stove   technology existing in early  1980*s models.    All  but  the
 smallest stove were  plate  steel  models.   The smallest stove was a
 cast iron model  with an inlet air control in the door.  Data from
new  stoves tested in 1984 and 1985  tend  to represent more recent
stove designs,   some of which were specifically  developed to meet
woodstove emissions  standards.    These  "third generation" stoves
also  show  significant increases in  emissions at burn  rates   of
less  than  40 percent (40%)  per  hour;   however,   other   factors

                                                           PAGE  18

-------
  60
  50
o>
^
irt
E
o
40
  30
CO
CO
LJ
   20
   10
 FIREBOX
 (ft3)
00.8
A 1.4
*2.0
 • 2.4
 • 3.5
STOVE
CODE
  BB
  CC
  DD
   2
  AA
             20      40       60      80      IOO      I20
                      %  FUEL  LOAD BURNED PER HOUR
                                  Figure 9
           Emission factors as a function of the  percent of fuel load
           burned per hour for second generation  RWC stoves with various
           firebox sizes.

-------
                                      WOODSTOVE DESIGN FACTORS




 appear   to   be  acting to  reduce emissions  below  the  baseline,

 especially   the mid-ranges of heat output  (Figure  10).   Firebox

 size,   based on volume alone,  appears to be the best first line

 screening   indicator  of possible  stove  performance  potential.

 Other   factors  which  are discussed  in other  sections  of  this

 report,  such as air supply systems and firebox shape,  appear to

 be capable  of further emissions reductions.



 2.  Firebox Shape - Firebox shape relates to the configuration of

 the walls,   floor and top of the firebox.   To a certain  degree,

 the  firebox shape controls the route of combustion air and  com-
                                   *
 bustion products through the stove, and is therefore closely tied

 with primary air supply and flue exit locations.  The most common

 firebox  shape  found on all types of woodstoves is an  elongated

 box,  designed to accomodate fuel logs either parallel or perpen-

 dicular to  the fuel loading door.  Width is generally the longest

 horizontal  dimension.   Height of the firebox varies,  but it  is

 not  uncommon  for  stoves to have roughly equal  dimensions  for

 height, width and depth.

    A  small  fraction  of stove designs  incorporate  barrel  or

 tubular  shapes.   One manufacturer of a round  stove  postulated

 that  a  round firebox increased combustion  temperatures,  since

 "reflected  heat from firebox walls was focused on the middle  of
                  *
 the  firebox."  This is not a documented phenomonon.   One  round

 stove design tested with a firebox volume of about 2.2 cubic foot

 showed average or higher emissions (Stove EE). It does not appear

 that there is any inherent advantage to round or spherical  fire-

boxes,  other than that they can minimize firebox surface area for



                                                          PAGE 19

-------
    60
                                                             Third  generation stoves
   50
    40
a
JC
a
|   30
a
(0
O
55
«2   20
UJ
    10
                                      FIREBOX
                                        (ft3)
                                       A 1.3
                                       « 1.3
                                       • 1.3
                                       • 1.4
                                       ©1.6
                                       D2.0
                                       A 2.1
                                       O2.3
                                       + 2.6
                                       H 1.4
                                         STOVE
                                          CODE
                                           M
                                           S
                                           Q
                                           F
                                           D
                                           00
                                           K
                                           B
                                           0
                                           C
                                       w/o 2nd air
                                           C.
                                        w/2nd air
                                           A1
                                           A2
                                           T
                                      Second
                                      generation
                                      stoves
               20
40
60
80
100
120
                                               140
                                               160
                             <*> FUEL LOAD BURNED PER HOUR
                                      Figure  10
                 Emission factors as a  function of the percent  of fuel
                 load burned per hour for recent design  RWC  stoves.

-------
                                      WOODSTOVE DESIGN FACTORS

 a  given  volume and therefore  reduce heat  transfer away from  the
 firebox.   This  would theoretically help  maintain higher temper-
 atures  and promote better combustion conditions in the firebox.
     Firebox height shows the  most variation among stove designs,
 due to  the. variety of flue gas baffling  and venting  configura-
 tions.    A baffling plate,  extending from the rear of the  stove
 upward  at a slight angle,  is  common in many stove designs and is
 often the effective firebox ceiling (Figure 11).  The addition of
 a  baffle was considered a significant improvement at the time it
 was introduced, with this design feature being considered "second
 generation"  technology.   Sloping sides or walls on  a  firebox,
 either  vertically or horizontally, are also sometimes used.
     Stoves  which  have demonstrated low  levels  of  particulate
 emissions  were  reviewed for  firebox shape to determine  if  one
 shape or  configuration was common to the best models.   While  no
 universal  shape  or  height-width-depth ratio is  found  on  the
 stoves  with the best emissions performance,  several models with
 low  emissions had relatively  short heights in relation to  width
 and depth (Figure C-6 in Appendix C).
     A  "squat" shape,  with a basic aspect ratio of about  1:0.5
 (Figure 12),  representing the ratio of width or depth to height,
 is  theorized  to promote complete combustion by  minimizing  the
 firebox size and fuel load while maximizing air/fuel mixing.  The
 low height (10 to 12 inches)  may be the most significant factor.
     A  "door wash" air delivery system was common to  all  three
 "squat"  stove models.   This design brings air in through a slot
at the top of the fuel loading door and is believed to allow good

                                                          PAGE 20

-------
          Figure 11
Single Rear Baffle Design

-------
        Figure 12
LOW FIREBOX HEIGHT DESIGNS

-------
                                       WOODSTOVE  DESIGN FACTORS

 mixing of fresh air with fuel gases.    A wide inlet  opening maxi-
 mizes fresh air supplies in the combustion zone.   The cooler and
 more dense inlet air is thought to drop into  the  primary   combus-
 tion zone.   It appears that this  feature, intended  to keep stove
 door glass clean, contributes to efficient combustion.  Figure C-
 7   in  Appendix C illustrates low  emission rates  are effected  by
 air inlet location for several stoves  at all  heat output  rates.

 3.    Pathway of Combustion Products -  The third  category  of fire-
 box  geometry is the flow pattern, or the pathway of  combustion
 gases.   Flow  patterns  are a function of the firebox  shape  and
 combustion air system and can vary significantly  with burn rates,
 gas  temperatures,  or  the draft  generated  by   stove operation
 and/or  installation.  Draft is a weak  force;  however,  since it is
 the  driving force for all gas flow through the  system,   it is  a
 domninant  factor in forming flow  patterns.    Observations indi-
 cate  the  magnitude and direction of eddies and swirls  within  the
 combustion   zones can be changed dramatically by  small  changes in
 burn  rate,   temperature distribution in the firebox,   and/or fuel
 load  configuration.    It should be noted that these   factors,  as
 well  as fuel  load geometry,  change throughout the cycle.   Actions
 such  as  opening  and closing the stove loading door  or  even minor
 adjustments   to  inlet air control  settings have been observed  to
 affect  flow  patterns as evidence  by flame patterns   and   changed
 fuel  load burning patterns.
     Combustion   efficiency and heat transfer efficiency  are both
strongly  affected by flow patterns, which influence  gas residence
times,  temperatures,   and air/fuel mixing.    From a  theoretical

                                                           PAGE 21

-------
                                      WOODSTOVE DESIGN FACTORS





standpoint,  combustion  efficiency  will be optimized  by  main-



taining  elevated temperatures with an adequate oxygen supply for



as  long as possible.   Flow patterns in the firebox  are  signi-



ficant in that they control  gas residence times and mixing rates



in  the hottest regions of the stove.   Flow patterns can be con-



trolled or directed in several ways:





  *  Gas  recirculation  (backmixing) - Some  stove  models  have



     designs which enhance the recirculation of combustion  gases



     in the firebox.   In this approach, combustion air enters the



     firebox,  mixes   with  hot fuel  gases and makes one or  more



     passes   through a swirl or eddy which supplies air  to  the



     primary combustion zone.   These gases complete their travel



     through the  swirl pattern within the firebox  before exiting



     the  stove.    This "recirculation" promotes turbulent mixing



     of  the hot  aged combustion gases with fresh combustion  air



     and fuel gas.    In  some stoves  (stoves B,  R,   and others),



     the  shape  of  flames formed  in the firebox  indicate  that



     recirculation  to some extent does appear to occur,   although



     this  is subject   to burn conditions and draft.





     While most stoves do achieve limited recirculation, measure-



     ment  and verification of this  phenomenon is difficult,  even



     when   visible  flames are present.   Monitoring the  direction



     of combustion  gases is difficult,  especially under the  low



     velocity and variable conditions of low-fire burn rates.



  *   Long   flame  path - This design attains long residence  times



     at  high temperatures by an  extended single path through  the



     stove.    High  temperatures  must be maintained along the en-





                                                          PAGE 22

-------
       Figure 13
LOW FIREBOX EXIT DESIGN

-------
                                      WOODSTOVE  DESIGN  FACTORS

      tire   path   for  this  concept  to be  effective   at   increasing
      combustion  efficiency.    Stoves using  this  design  have   been
      observed  to be ineffective  due to  the inability to maintain
      adequate  gas  temperatures far from the primary   combustion
      zone.

      All  of   the non-catalytic stoves   which  have demonstrated
 relatively   low average emissions  (less than 15 grams/hour)  at
 relatively  low burn rates  have primary air  introduced high in the
 front of the firebox  and a firebox breach in the front  section of
 the top of  the firebox  (e.g.  Stoves B,  C,  M,  O, S,  T, and Q).
 While these designs have other differences  (shape,  size, with or
 without  secondary  air supplies,  etc.) the common trait of  the
 combustion  gas  flow  pattern appears to  be  a significant  factor.
 It does promote  good  mixing of the air,  fuel,   and burning gases
 in the combustion zone.
      Most   stove designs have a  gas flow configuration in  which
 combustion  products  exit  at the top of  the stove.   Some  stoves
 however,  use  a low  exit,  typically at the bottom rear  of  the
 firebox,  to provide  a longer residence  time in  the stove (inten-
 ded   to  enhance either combustion or heat  exchange  efficiencies
 (Figure 13).  These designs use the rear of the  stove for primary
 heat  transfer rather than the top.   A  limited  number  of  stoves
 have been tested which use low rear exit designs.
     The low rear exit design poses some potential  problems on   a
 theoretical  basis.  Relatively poor test results from  stove  models
with  this feature seem to bear this out,   although other  design
features  may have contributed.   Stoves E  and Y are examples of

                                                          PAGE 23

-------
                                       WOODSTOVE DESIGN FACTORS

 stoves with low exhaust gas exits from the firebox.    Theoretical
 problems include:

   *  Hot  combustion  gases  are  buoyant  and  rise  within   the
      firebox.    In order for these gases to exit the stove,   they
      must  either  cool and drop to the exit or be pulled down  to
      the  exit   by static pressure in the flue (draft)   which  is
      great enough  to overcome the buoyant pressure of the firebox
      gases.
   *  A low exit from the firebox can make the  stove  difficult  to
      start,  as a  strong draft is required to  draw hot gases  down
      to  the exit.    In these cases/  a bypass damper is  usually
      required   for  startup and for refueling.    Without a bypass
      damper,  flame  and  smoke spillage into  the room can occur
      when  the  stove door is open.   An open  door,   providing  a
      higher  opening  than  a  low exit,  is a  pathway  of   less
      resistance to the flow of hot combustion gases,  unless  an
      extremely   low flue pressure (high draft)  condition  in   the
      flue exists.
a. Baffling
      Baffling   is a feature which is significant in  defining   the
size  and  height  of  the  firebox as well as the  flow of  combustion
gases   through   the stove.    A baffle which is low and  covers  a
large   portion  of  the  firebox  can  help  improve  combustion
efficiencies by  lowering the height of the firebox and  restrict-
ing   the  size   of the fuel  load.    This can also be an important
feature   in enhancing the heat exchange capabilities of a stove.

                                                           PAGE 24

-------
                                      WOODSTOVE DESIGN FACTORS

 However,  the term  "baffle"  in the woodstove  industry  incorporates
 a  wide  range of designs with an equally wide  range of  effective-
 ness.
    The   most  common  baffling arrangement  is  a  sloping  plate
 extending from the rear of the firebox forward and upward  at  a
 slight angle as shown in Figure 11.   Combustion gases are forced
 to travel  around  the  baffle before exiting  the   stove.   The
 length  of  the baffle will affect the  "route"  the  gases  must
 make;   a longer baffle will increase the combustion gas pathway
 and  increase  gas  residence  time.   As  mentioned  previously,
 another   significant  feature  of a baffle is that  is  acts  to
 impose on or limit firebox ceiling height.
     Multiple baffling is used by some manufacturers,  usually to
 increase  heat exchange efficency (Figure 14).  In this configura-
 tion,  combustion  gases  may  be brought forward  by  the  first
 baffle,   routed  again  to  the rear of the stove  by  a   second
 baffle,   forward   by  a third and to the rear again by  the  top
 surface of the stove. This lengthy routing of flue gases can help
 heat transfer to the room,  although OMNI does not have test data
 on this design.  A higher resistance to flow is usually generated
 by  multiple baffling,  which may necessitate a bypass route  for
 startup  and fuel loading.   Higher heat transfer efficiency  and
 closer  wall   clearances (determined by a stove  safety  testing
 laboratory)    are  often a benefit of the resulting   lower  flue
 temperatures.  However,  lower flue temperatures can contribute to
 increased creosote deposition in the flue.
     The  baffle  plate  is,  in effect,  the roof of the  primary
firebox.   The height and shape of the baffle can therefore be an
                                                          PAGE  25

-------
        Figure  14
MULTIPLE BAFFLING DESIGN


-------
                                      WOODSTOVE DESIGN FACTORS




 important  feature  in stove design.



 b.  Combustion Chambers - Primary and Secondary


     The fueling and firing of a woodstove occurs  in the  primary

 combustion  chamber.  The primary combustion chamber is the fire-

 box,  where fuel is loaded and burned.   Virtually all combustion

 processes  and  the release of energy from the fuel  takes  place

 here,  at  high or low burning conditions.   A number  of  stoves

 currently  on the market advertise secondary combustion features.

 These  include designs with secondary air supplies and/or  secon-

 dary combustion chambers.   Secondary combustion,  in theory, pro-
                                          »
 duces  more  complete burning of incomplete  combustion  products

 ("smoke") before these products leave the stove.  This can theor-

 etically  be accomplished by introducing fresh air into  the  hot

 gases  at the exit of the primary combustion chamber.  Combustion

 gases  leaving the primary combustion chamber often have  reduced

 oxygen concentrations so that with the introduction of additional


 "secondary air",  additional combustion can  occur.  However,  in

 order to prevent quenching of the combustion process by cold air,

 secondary air must be preheated before entering the gas stream.

     In a technical sense, a  combustion chamber is not a second-

 ary  combustion chamber unless additional oxidation of incomplete

 combustion  products is induced and subsequent release of  energy

occurs.    Under this definition,  very few non-catalytic   stoves

have  true  secondary    combustion systems.   Most stove designs

are  not capable of maintaining the temperature of gases  exiting

the  primary chamber at low and medium burn rates at levels  high

enough to support  secondary combustion.  Test data indicate that



                                                          PAGE 26

-------
                        •  .             WOODSTOVE DESIGN FACTORS

 true secondary combustion on most stoves only occurs at high burn
 rates when gas temperatures are very high. Most secondary combus-
 tion  chambers seem to act simply as a baffling chamber  or  heat
 exchanger  surface  when the stove is operated at low to  medium-
 high burn rates.  This can be seen from the fact that stoves with
 secondary  combustion  chambers still exhibit  significantly  in-
 creased  emission rates at burn rates of less than 40%  of  their
 fuel load per hour (Stoves B,  Cf  D, and T).  Stove R has demon-
 strated  an effective secondary combustion chamber system (Figure
 C-8  in Appendix C).
      The   key difficulty appears to be introducing secondary  air
 at a point where combustion gases are at temperatures high enough
 to   promote additional combustion (typically  well above 1100  de-
 grees F) .    Most models have secondary air inlets at the exit  of
 the   primary firebox  where gases begin to cool below the  temper-
 tures  necessary  for the combustion of hydrocarbons  and  carbon
 monoxide.

 c. Downdraft Combustion
      The   principle of a  downdraft stove is to draw burning gases
 down   through  a  bed of coals to promote more  complete  combustion
 of the volatile  compounds generated in the main fuel mass (Figure
 15).   Theoretically,   the  combustion zone of the downdraft stove
 is  confined to  the lower coal  bed areas of the  fuel  mass.    By
drawing the  hot  combustion  gases down through the coals,  the fuel
mass, other  than  that  in  close  proximity to the coal bed, remains
 relatively   cool.   This  limits,  or  more closely regulates,   the
 release  of  fuel  gases   from  the fuel mass   and  promotes  more

                                                           PAGE 27

-------
Air Inlet
    Goal Bed
                     Figure 15
              Downdraft Woodstove

-------
                                      WOODSTOVE DESIGN FACTORS








complete and even combustion  conditions  in the primary combustion



zone.  In practice,  however, downdraft  stoves have not exhibited



significant  performance or emission improvements over the  base-



line updraft type of stove.   Figures C-9 and C-10 in Appendix  C



show  the  relative emission  rates of downdraft and updraft  type



stoves.
                                                          PAGE 28

-------
                                     WOODSTOVE  DESIGN  FACTORS
 C.    COMBUSTION AIR -  DISTRIBUTION,  CONDITIONING AND CONTROL

      1.   Primary Air   - Air  supply  systems  include  all  devices
 affecting  the  volume  and  distribution of  inlet  combustion  air.
 Typically  combustion   air enters  the stove  through one  or  more
 openings,   often located symmetrically, through which flow can be
 regulated  and  controlled.    It also includes any  intrinsic  lea-
 kage,  intended or  not,  through door seals, around air controls,
 and   through cracks in  the stove.
      Primary  air refers to the main flow  of air into the primary
 combustion  zone or  firebox.   This air controls the overall  rate
 of  fuel  consumption   and the heat output of  the  stove.   Many
 stoves use  a single air inlet,  although a number of other confi-
 gurations are also  used.   Common air inlet configurations include
 dual  (symmetrical)  inlets on the door(s)  or sides of the  stove,
 at the rear of  the  stove/  at the top or bottom of the  door  and
 from  underneath  the floor of the firebox.   The  upper  firebox
 inlet,  or door glass air  wash system,  has seen increasing popu-
 larity  on stoves,  as  it  helps prevent the accumulation of creo-
 sote  on glass windows.   Most important,  it promotes mixing  of
 fresh air with the burning and aged combustion gases.
    The  amount of air entering the firebox is a function of  the
area  of  "free flow" through air ports and the  draft  (negative
pressure)   drawing air into the stove.   The flow of air is  gen-
erally controlled by a device which can decrease or increase  the
free flow area available for air to pass through.  Common designs
 include  a sliding plate placed over the primary air port,  "spin

                                                           PAGE 29

-------
                                      WOODSTOVE DESIGN FACTORS

 drafts," butterfly valves and flapper plates,   with or without   a
 thermostatic control.   The pathway of inlet  air is controlled  by
 the location of entry and temperature.
      The  location  of the air entrance into  the firebox  can  be
 important  to the burning characteristics of  a stove for  several
 reasons.  If air is brought directly into the  firebox too  low,  it
 can descend in the firebox and "tunnel" into  and under  the  coals
 due  to  its cooler temperature and higher density relative to the
 hotter aged gases in the firebox.    These conditions can lead  to
 pockets  of oxygen depletion in the coal bed and fuel mass.    When
 this  occurs,   carbon monoxide concentrations  increase in the flue
 gases. For example, measurements of carbon monoxide in flue  gases
 during   non-tunnel  conditions  in a baseline  stove  were  1.25%.
 During   a  period when tunneling was observed  in the  fuel  mass,
 concentrations  increased to 3.50 - 4.60 %.   In addition,  as raw
 wood  is  heated,   fuel gases are volatilized and driven off   from
 the fuel  mass.    Incomplete combustions of these gases contribute
 to particulate  emissions loading.
     Air   should  be brought into the firebox so  that  sufficient
oxygen is  mixed  into aging combustion gases, and that an adequate
amount  of air  reaches the coal bed.   Most stoves showing   good
performance   (averaging less than  15 gram/hour)  at  low to  medium
burn rates have  air  inlets elevated in the firebox  (Stoves A,  B,
C, D, M,  S, T).
     2.   Underfire  Air - While most stove designs introduce   com-
bustion   air through the walls or  door of the  firebox,  underfire
air  systems bring  air into the firebox through a grid  or   grate

                                                           PAGE 30

-------
                                      WOODSTOVE  DESIGN FACTORS

 under  the  fuel  mass.    This  design is  typical   of   combination
 coal/wood burners.   When  wood  is  fired from below, heat  released
 by  combustion  rises  through  the fuel load,   releasing excessive
 amounts of the  volatile  components of the wood.    However,  obser-
 vations  in  the OMNI lab  show  that low flue oxygen   levels   are
 typical  of underfire designs.   This results in poor combustion,
 high  particulate  and  carbon  monoxide emissions  and is  char-
 acterized  by  very low  oxygen  in  the flue.    For this  reason,
 combination coal/wood burners,  or any woodstove using underfire
 air,   have  performed poorly  in emission  tests.    Figure  C-ll   in
 Appendix   C  illustrates   the effect of   reducing  underfire   air
 conditions  in   a  woodstove  air supply   system.   Onderfire   air
 designs  show  performance similar to or  worse  than   those  which
 introduce  air   very   low  in  the   firebox,   for  similar  reasons
 (Stoves El,  E2,  and FF).
    One  of  the  most  effective  primary air supply  systems appears
 to be  a wide air  slot  at the top of the firebox (Stoves A,  B,  C,
 D,  M,  O,  Q,  S,  and T  ).   This  approach was  devised to keep glass
 windows   clean by preventing the deposit  of  creosote  or   carbona-
 ceous material.    Air  is typically drawn  into the  firebox through
 a slot  extending  across and directly above the  stove  window.   The
 air  "washes" down  the glass,   providing  an  air curtain   to keep
 combustion gases  away from the  window.  When  sized appropriately,
 these systems are  quite effective  at  keeping  glass windows  clean,
 even at relatively  low burn rates.
     There are several reasons  for  the effectiveness  of door wash
designs   in   reducing pollutant emissions.    The introduction   of
primary  air  high  in the  firebox  (at  the top of the   door)  will

                                                            PAGE 31

-------
                                      WOODSTOVE DESIGN FACTORS

 generally  allow good heating and mixing/   preventing pockets   of
 oxygen   depletion  in  the  coal  bed  and  fuel   mass.    Also,
 introducing  air  through  a slot  extending   across   the   stove/
 combustion  air  can  be mixed more readily with hot   aging fuel
 gases.  Most door wash stove designs also  have the  firebox  breach
 toward the front of the stove,  so that  the aging gases mix with
 some  of  the  high-oxygen inlet air before  exiting   the   stove/
 promoting  additional combustion if temperatures are  sufficiently
 high.    The door wash designs also appear  to  promote  a circular,
 "swirl"   type gas flow pattern,  which promotes efficient combus-
 tion through greater turbulence and longer residence  times.
     3.    Secondary  Air - Secondary air  is a  term describing   air
 added  to hot combustion gases which have left the primary combus-
 tion  zone.    Although this term is used to describe  a variety of
 designs,   few  stoves  have what can  be  technically considered
 effective  secondary air systems, especially at medium to low burn
 rates.   What  is  called secondary air  is often simply an  addi-
 tional air  supply.
     A well  designed secondary air system  allows preheated  air to
mix  well with hot aging combustion gases.   To be effective, this
air  should   be  brought into  an area  where  rapid  and  thorough
mixing is possible.    The point of injection  needs  to be as close
as  possible  to  areas where the highest  temperatures  occur  at  any
burn  rate.    Secondary air ports  on several  low emission   stoves
introduce preheated air at the top of  the  firebox,  where   mixing
and combustion can  occur before gas  temperatures drop (Stoves  C2-
2,  C2-3,  M,  R).   At low burn rates,   most  secondary combustion

                                                            PAGE 32

-------
                                      WOODSTOVE  DESIGN FACTORS

 systems  lose  effectiveness   due   to low   gas   temperatures   and
 turbulence.
      Effective   secondary  air  systems are  possible  on  woodstoves
 (Stoves B,   C2-2,   C2-3,   M, and R).   These stoves  introduce pre-
 heated   air,  directly  into the upper flame zone of  the   primary
 firebox.   The   air is  diffused across a wide area  to allow  com-
 plete mixing.   Data comparisons of  various secondary  air   systems
 are  presented in Figures C-12  through C-14 in Appendix C.
      Secondary   air volumes are low in comparison to  primary   air
 volumes  and  may   be regulated together or separately from   the
 primary  air.    A feature  which blocks secondary air  completely
 below  a  certain primary  air  setting may  be desirable to  reduce
 high excess  air conditions when the  secondary  air  is not  effec-
 tively   improving combustion efficiences.    This is a feature   on
 Stove M.
      4.   High Minimum Burn Rates -  One approach to  the problem of
 low   combustion  temperatures and efficiency at  low  burn rates   is
 to   always maintain  a moderate to high burn rate by not allowing
 the  air controls to  be closed below a certain point.  As presented
 previously, non-catalytic woodstoves   have lower combustion effi-
 ciencies  and  higher particulate emission rates at firing rates
 below about 40%  of the fuel level burned per hour.    This  is   due
 to lower combustion  temperatures and  poorer mixing  of gases.    In
 addition,  a larger  firebox (and hence a larger fuel  load),  will
 result  in  higher emissions at any low burn rate,  as described
 previously.
     Current  emission  testing requirements  prescribed   by   the
Oregon  DEQ  and in  the proposed ASTH test  procedure  (8)   specify

                                                           PAGE 33

-------
                                      WOODSTOVE DESIGN FACTORS

 that an appliance be tested across the range of stove heat  output
 capacity.  If  a  stove is not  capable of  passing  an   emission
 standard at the prescribed lower burn rates/  a minimum  burn  rate
 which  is  high enough to provide good combustion  conditions   and
 reduced emissions can be established.  This  is accomplished by an
 air   control  stop which prevents closure  beyond  an   established
 amount.   Elevated minimum burn rates are  used to  reduce emission
 rates,   especially with larger firebox stove models (Stoves D,  F,
 and  G).  It is important to note that while  a tested  minimum  heat
 output  may be relatively high,  actual "in-use" heat   output   can
 be   reduced by firing the stove with  smaller fuel  charges.    This
 approach does reduce emissions based  on units of heat delivered.
      5.   Controls - Adusting  the air control allows  a greater  or
 lesser   opening for air entering the   firebox.   This adjustment
 controls  the rate at which fuel is consumed and the  heat   output
 of   the stove.    Air inlet controls are typically  either a  manual
 control  which  remains  in a  fixed position  or  a  thermostatic
 control  which  responds  to changes  in stove   or room ambient
 temperatures.
     A  variety of designs  are  used for manual controls.   Sliding
plates,  hinged  flaps,  butterfly valves and spindraft  controls are
most  common.    An adjustment  lever is usually  connected directly
to  the air  control  and is therefore  located near  the air   inlet.
Some designs  use connecting linkage between  the adjustment  lever
and  the  air  control,   permitting the control lever and  air inlet
to  be  separated.    This may be done  for convenience  and ease  of
operation, or  may help keep control levers cool.

                                                            PAGE 34

-------
                                      WOODSTOVE  DESIGN FACTORS

      Thermostatic  devices  are  often  used  on  woodstoves,   although
 they  are   more  common  on   catalytic  models.    Thermostatic   air
 controls typically use  a bimetallic  spring or coil  in conjunction
 with a manually  adjusted control.  The thermostat reduces the  air
 opening when  desired  temperatures are  achieved   and  opens   to
 provide increased air supplies  as  stove temperatures  drop.
 While  the  bimetallic coils  are  most common, oil-filled  sensors
 and  linearly expanding  rods  are  also used.
      When   fuel  is  added to  a  non-thermostatically controlled
 stove,  temperatures peak  approximately halfway through  the burn-
 ing   cycle  if the air  inlet control remains constant.   With  a
 thermostat,  however,   the thermostat begins to reduce the amount
 of   air entering the stove as  stove   temperatures   increase.    By
 this   time,  most  of the  fuel charge has  been  heated and is   in-
 volved  in combustion or is volatilizing fuel gases.    As tempera-
 tures   increase  and the thermostat  closes,  air to the   fire   is
 reduced,  and  the heated  fuel charge continues to  emit organic
 compounds.  In most cases  the  decreased air  supplied  by  the ther-
 mostat  is insufficient for complete combustion of  the fuel gases
 evolving from the  fuel mass.   Low excess air conditions  and  high
 emissions  result.  The apparent effect of this is  illustrated in
 Figure C-15 in Appendix C.  As a general rule,  it has been OMNI's
 experience  that a thermostatic device on a  non-catalytic  stove
will  cause  increased  emissions if compared to the   same  stove
without  a thermostat.   Data  from Stoves  E  and FF  exhibit  this
 charateristic.
                                                            PAGE  35

-------
                                       WOODSTOVE  DESIGN FACTORS









 D.   HEAT  EXCHANGE SYSTEMS





      The  overall heating performance  of a  woodstove  is based on



 two   parameters:   combustion efficiency and heat transfer  effi-



 ciency.   While  the efficient combustion of wood in a naturally



 drafted   appliance has shown to be a complex and difficult  task,



 efficient transfer of heat from hot flue gases to a heat transfer



 surface is relatively  well understood.   Many woodstove  designs



 under-utilize existing technology for  maximizing heat  transfer.



      The  heating  capabilities of a   woodstove are based on  the



 ability of the stove to transfer the heat released from the  fuel



 by  combustion  to the space to be heated.   To accomplish  this,



 heat  energy must be removed from the burned gases, passed through



 the   surfaces  of the stove and transferred into the   space  sur-



 rounding  the stove.   Heat transfer from a fluid (gas) to  solid



 surface and vice versa is a function of the velocity of the  gas,



 the geometry and area of the surface,  the emissivity  of the gas,



 and the temperature differential between the gas and the surface.



 Most  basic  stove designs use the combustion chamber  as  a  heat



 exchanger.  A  simple box stove design without baffling transfers



 heat from hot combustion gases to the  stove surfaces,   which then



 radiates  heat to the surrounding air.   On all stoves, the chim-



ney  connector  between the flue collar and chimney  is also  an



effective  radiating  surface.   A significant drawback of  this



general  technology  is  that by removing heat  from   the  entire



stove,  firebox included,  temperatures in the combustion chamber



are lowered,  which can reduce combustion efficiency.








                                                        PAGE 36

-------
                                       WOODSTOVE DESIGN FACTORS

      For any stove design,  heat transfer efficiency will  usually
 decrease  as  the burn rate increases.   This is due to the  de-
 creased  residence time of the hot combustion gases in the  stove
 due to higher flow rate/ even though higher temperature differen-
 tials exist.  There does not appear to be any way to prevent this
 loss  in  efficiency  at  higher burn  rates  without  increasing
 residence time or radiating surface area. The lower heat transfer
 efficiences  at high burns may be compensated for by higher   com-
 bustion efficiencies, so that overall efficiency may remain  rela-
 tively constant.
      The  design   considerations  mentioned  above  help  improve
 transfer  of heat from combustion gases to the  stove  body,  and
 high  surface  area on the stove will increase the heat radiated
 into the room.    Some stoves also rely on forced convective   heat
 transfer to the space being heated by using a fan to force air up
 the   back across  the top,   and/or across  the sides of the  stove.
 Fans  are  often  used on fireplace inserts,   due to  the  reduced
 stove   surface  area available for radiating heat into  the  room.
 However,   it  has not been shown that a fan will increase  the net
 performance of  a  stove.    This  is  especially true if a fan  draws
 enough   additional  heat  away from the stove to cool  the  firebox
 temperature and thereby  decrease combustion efficiences.
     A   number  of   stoves use  a natural  drafted  convection  air
 system   to  draw  air up  past the sides of  the  stove.   In   this
approach,   an outer shell  or cabinet  (typically light gauge  metal
or   ceramic  tiles)   is  placed  around the  firebox.   Heated  air
between   the  stove and  the shell rises and  produces  a  thermal

                                                         PAGE 37

-------
                                      WOODSTOVE  DESIGN  FACTORS

 siphon effect,   drawing  cooler  air  in at the bottom  while  venting
 heated  air  at  the  top.   This  method also  usually allows   closer
 clearance  from the stove to combustible surfaces when  tested   to
 the  safety standards used in the  industry.
     While   it   is   desirable to  obtain as  much  usable heat   as
 possible from   a stove,  some  heat must be sacrificed  to   ensure
 proper  operating conditions.   If too much heat is  removed from
 exhaust  gases,  flue temperatures may be inadequate to maintain
 adequate draft  pressure.  The lower the draft,   the  less air will
 be drawn into the stove  and the slower the  wood  will burn  at  any
 given  setting.
     Maintaining  a  minimum stack temperature also   reduces  the
 chance  of   condensing water in the flue.   With  an   average flue
 moisture content of  12%,  condensation of  moisture  will occur  at
 about  120 degrees  F.   At a moisture content of  25%,  which   is
 common   in woodstove  flues for short periods,   condensation will
 occur  at 150 degrees F.   This potential problem is of   greater
 concern  on  catalytic stoves,  where stack gas  temperatures  and
 velocities are typically lower.   Moisture  condensation can be  a
 nuisance problem by draining down the wall  of the chimney  into  or
 onto the stove.   If moisture condensation  in the flue  is  a chro-
 nic  problem,  corrosion  of the metal chimney   connector,  metal
 chimney  and  even the stove itself may be  greatly   accelerated.
This  problem can also be exacerbated by freezing  conditions   in
 flue segments located in outdoor environments.
                                                        PAGE 38

-------
                                       WOODSTOVE DESIGN FACTORS

 E.   CONSTRUCTION MATERIALS

      The  materials used in the construction  of a   stove  can  have
 a significant impact on the performance of  the  device.  There are
 two primary functions in a woodstove:    releasing   heat from  the
 fuel  and recovering heat from combustion gases before  they   exit
 the  stove.   In efforts to maximize the efficiency of  these  two
 functions,  opposite tasks must be achieved. High temperatures are
 essential  for  complete combustion,   requiring that heat be  re-
 tained in the firebox as much  as  possible.  Efficient transfer of
 heat out of hot  flue gases/ however,   requires good conductivity
 of  heat away from the flue gases.
      A  traditional woodstove, constructed   entirely with single
 walls of plate steel or cast iron,  accomplishes much of  its  heat
 transfer by radiating heat from the firebox,  which  serves as  both
 combustion  chamber  and heat exchanger.   The resulting lower fire-
 box  temperatures  make efficient combustion more difficult.   The
 use   of  a  full baffle in the  roof of  the firebox does  help  in-
 crease   firebox temperatures by reducing the  temperature  differ-
 ential   of  the top  of the firebox  (the  baffle  plate) as  shown in
 Figure  16.
      Most   stoves currently use a plate  steel or cast iron  fire-
box.  Plate  steel is most common, primarily because the equipment
 required  to  form, shape and join the metal is relatively  inexpen-
sive  compared  to  the cost of casting.   Steel  stoves typically use
 1/4"  plate   for  the firebox  and 3/16"  plate   for   other  parts.
Cast  stoves  offer greater potential for  unusual shapes  and forms,
and   parts can be formed rapidly and economically once  a  design

                                                        PAGE  39

-------
          Figure  16
HEAT RETENTION EFFECTS OF BAFFLE
                 300" F
        80
rp     A
            500 F
            900*F

-------
                                       WOODSTOVE DESIGN FACTORS

 mold has been established.   Parts are usually joined with  screws
 and clips and seams are caulked.
      Maintaining high firebox temperature for  any given burn  rate
 can  be accomplished by utilizing materials which have a low  heat
 transfer coefficient.  Recent work by Fuentes  and Hodas (9) iden-
 tifies the insulating effectiveness and relative costs of various
 ceramic materials/   including firebrick.    As  a practical matter,
 any material  or design which reduces heat transfer away from  the
 firebox is desirable.   A more efficient  refractory material  will
 achieve lower heat  transfer (conductivity)  coefficent for a given
 thickness.  Less efficient insulators will  require greater  thick-
 ness to attain the  same resistence to heat  loss.
      It should be noted that although it  is desirable to maintain
 high  temperatures  in the primary combustion zone,  there can be
 undesirable   effects  if temperatures get  too  high.    Excessive
 temperatures   cause the fuel mass to heat up too  fast,-  accelera-
 ting  the  release of the volatile fuel gases.    without the  addi-
 tion  of large amounts of air,  these volatiles  burn inefficiently,
producing high pollutant emission rates.  For  example,  an experi-
mental  stove  tested  at  OMNI  had a  highly   insulated primary
combustion  chamber   which maintained temperatures  at  1820-1980
degrees F.    At  the  end of each  burn cycle  flue gas carbon  mono-
xide  concentrations were in the range of 0.1% to 0.2%,   with 10%
oxygen.  However, each time additional test fuel  was  added  to the
coal  bed,  carbon monoxide levels exceeded 10% with oxygen  less
than  2%.   Dense   smoke was also observed  in  flue  gases  during
these periods.
                                                         PAGE  40

-------
                                       WOODSTOVE  DESIGN FACTORS

      While the  use  of  refractory  material  to  insulate  the  firebox
 is recommended  from a  theoretical   standpoint,  verification   of
 how  much   is   necessary  has  not  been  attempted  on   a   controlled
 basis.   Tests   on   the   addition of refractory  material   to   an
 existing   stove  design   have  been  performed   (Figure C-16   in
 Appendix C).    However,   no tests have been performed  to separate
 the  effects  of reducing  the volume of  the  firebox  and  the
 effects of the  insulating value of the refractory.   Most   stoves
 with  a refractory lining  incorporate a standard  firebrick  used  by
 masons in  building  fireplaces.  These  bricks are nominally 4.5  by
 9  inches and typically 1.25 inches thick,  although  2.5 inch thick
 bricks are available.
      Heat  exchange  materials  usually consist of  the material used
 in  the construction of the stove.   Most  stove  designs use  the
 walls  and top  of the stove as  radiating surfaces,   and virtually
 all radiating metal surfaces  are  painted black to increase emis-
 sivity.    Some  stove designs  utilize fins  to increase  the  radiat-
 ing and conductive  surface area.
     Some  stoves advertise "high  mass" heat storage or efficiency
 characteristics.  These stoves  work  on the principle of heating a
 large  mass  of material  during the  firing period and   then  re-
 leasing  the  heat  slowly over a  longer period  of time.   This
 concept  holds potential  as an  emission reduction method only   if
 combined  with both an air control system which  will   maintain   a
 high minimum fire setting and an  appropriate heat transfer system
 to  capture  the produced heat.    The  principle  of  using a large
mass  to   store heat is that  heat  absorbed by the material at   a
high firing rate is radiated  to the  living space at a  slower rate

                                                        PAGE 41

-------
                                       WOODSTOVE DESIGN FACTORS

 over  a longer period of time.  This allows a stove  to  be  fired  at
 a   higher rate than would be comfortable with a similar steel   or
 cast  stove.   The high  firing rate will  usually result in a  lower
 pollutant emission rate.   However, if the air control can be set
 to  maintain  a low burn rate,  resulting  in higher particulate and
 carbon monoxide emission rates,  no emission reduction benefit  is
 realized.  While a high mass stove will  radiate stored heat  for a
 longer period than a lighter stove,  it   will also  take a longer
 time  for  the stove mass to come up to temperature.
      A high  mass/high heat capacity stove allows a  higher firing
 rate   over a longer period of time than  would be acceptable  in   a
 lower  mass stove.   Heat is released from the high mass stove at a
 slower  rate than  it is accumulated during the firing  period and
 then  releases it longer after the firing period is  finished.    As
mentioned  previously,   no emission reduction benefit  is  seen un-
less a  high  firing rate is used at all times.    Heat output  from
the  stove   could  be regulated by the frequency of  firings,  and
continuous firing  may not  be necessary.
     A  variety  of materials are used for  heat storage and/or
decorative purposes.    Ceramic,   stone and refractory-type mater-
ials  are  frequently  used for  their low cost and  high  specific
heat  (approximately twice  that of steel).
                                                        PAGE  42

-------
                                      WOODSTOVE  DESIGN FACTORS
 IV.   OTHER  NON-CATALYTIC TECHNOLOGIES

 A.   Pellet  Fuel Stoves
      Efficient  combustion   is difficult to accomplish in  tradi-
 tional  wood  stoves at low  heat outputs due  to  low  combustion
 temperatures  in the firebox,  poor air/fuel mixing,  and changing
 combustion  conditions.  If air can be moved into the firebox in a
 more concentrated and turbulent manner than relying on the draft
 generated by  hot flue .gases,  higher temperatures and more  effi-
 cient combustion can be achieved.   Mechanical draft systems,  in
 which  air  is forced or drawn into the firebox,  are used on many
 wood furnaces  and  all types of  small,  mid- and  large  scale
 boilers.   This  technology  has seen virtually no application  to
 residential wood stoves, however.
     A forced draft system moves air into the firebox under posi-
 tive pressure;  higher pressures exist in the stove than the sur-
 rounding  air.   Induced draft systems draw exhaust gases out  of
 the firebox,  creating negative firebox pressures.   Both systems
 use  fans to move the air and gases,  hence the  term  mechanical
draft.   (Natural  draft  stoves  are  more  properly  "thermally
drafted").   The  use  of a mechanical draft allows much  greater
control  of combustion air flows and patterns,  resulting in  im-
proved combustion conditions.
     Both  the induced and forced  mechanical draft systems  have
advantages  and  disadvantages.   Forced  draft  systems  require
                                                           PAGE 43

-------
                                       WOODSTOVE DESIGN FACTORS

 higher  quality  construction methods to  prevent   leaks  through
 doors and other openings but move clean air through the blower.
      Induced  draft  systems  do not require a  perfectly  sealed
 system, as any leaks will simply draw air into the  stove or flue.
 They must move hot and sometimes dirty gases,  requiring more care
 in  selection and maintenance of fans and  motors.   In  addition,
 mechanical  draft stoves will not operate without electric  power
 to  drive  the fans,  so these units can not be utilized  during
 power outages.
      Among wood burning appliances currently available, virtually
 all   mechanical draft stoves are designed to burn pelletized wood
 fuel.  Pellets are typically 1/4 inch to 3/8 inch in diameter and
 about  1/2 inch in length.    The composition varies among  manu-
 facturers,  but  is primarily sawdust and chips from forest  pro-
 ducts operations.    Some pellets are composed  of wood  only, while
 others contain more bark and debris.   Most are formed under heat
 and  high  pressure,   and most use no binder.   Heat  content of the
 pellets  are typically 8750  to 9200 Btu/lb, with a moisture con-
 tent  of 6-10  percent.    Pelletized  fuel has   been successfully
 used   as  a substitute for coal in many small boiler applications
which have traditionally used coal.
      The   primary  advantage  of using pellets in residential  com-
bustors   is  the  ability to control the amount  of fuel  involved in
combustion  at  any  time.   Air and fuel feed rates  can  both  be
controlled,   allowing  optimized combustion conditions.    Pellets
can   be fed  at  a constant rate into a combustion zone   maintained
at  high  temperatures and high turbulence by a forced  or  induced
draft.

                                                            PAGE  44

-------
                                      WOODSTOVE  DESIGN  FACTORS








      The  mass  of  fuel  involved  in  combustion  at  any  time  is  very



 small,  while   oxygen  supplies  and turbulence are  high.   Pellet



 stoves  can operate  under steady state  conditions  as a  continuous



 process,   rather  than  the semi-batch process  of  burning cordwood.



 The  small  mass  of fuel burning  at  any given time promotes  stable



 conditions,    which  allows  more   fficient   combustion.   Pellet



 stoves have shown to have efficiences and  emissions  comparable  or



 better than catalytic  cordwood  stoves (Stoves U, and X).   Results



 are  illustrated in Figures C-17 and C-18 in Appendix C.



      On  all existing pellet stove designs,   fuel  is stored  in   a



 hopper and moved  into the combustion chamber/firebox with  a  moto-



 rized screw   auger.   The feed rate is controlled by a variable



 speed or  time-on switch.  Pellets are  pushed or dropped   into   a



 small cup-shaped  tray which is  surrounded  by  combustion air  inlet



 jets,  creating an intense burn region.  Blower  speed can  also  be



 varied;  some stoves combine air and fuel  feed rates while others



 offer  combinations of independent  fan  and fuel  feed rates,  and



 continuous  or  intermitant operation.  Most  pellet  stoves use   a



 refractory  lined firebox in which the  pellet "cup"  and air  ring



 are located.  Gases are then vented through heat exchange  baffles



 or chambers similar to conventional woodstoves.   Several  models



 of add-on pellet hopper/burner devices  are also  available,  which



mount  on  most woodstoves.   The  performance of a pellet  hopper



device on two stoves is shown in Figure C-19  in  Appendix C.



     Exhaust  gas temperatures are  generally  quite low  on  pellet



stoves,  indicating high heat transfer  efficiency.    Temperatures
                                                           PAGE  45

-------
                                       WOODSTOVE DESIGN FACTORS

 as  low as 150 degrees have been recorded on several  devices   at
 the OMNI laboratory.
 B.  Densified Fuel
      While  pelletized fuel requires  a special  combustor  and fans
 to move combustion air and exhaust, other processed  and densified
 wood waste logs have also shown lower  emissions  in    traditional
 stoves.  Limited testing on densified  wood  residue  "logs"  showed
 approximately  one half the particulate emissions seen from burn-
 ing  a  comparable mass of cordwood  in the same  appliance  at simi-
 lar heat outputs.   Stove runs  Z2 are densified fuel,  compared to
 Zl  runs  using Douglas Fir,  fuel.   Figure  C-20   in  Appendix C
 illustrates   these  data.   It   appears that this is due  to   the
 slower  evolution of fuel gases  because of the lower  surface area
 per   unit  mass.    Fuel moisture is generally quite low (about  8%)
 in the  densified fuel.

 C.   Mechanically Drafted Cordwood Stoves
      Very  few   mechanically drafted  residential  wood  stoves   are
 currently  available.   Most mechanically drafted units  designed to
 burn  cordwood are furnaces,  hot water heaters  or central heating
 systems.   Some  testing has been done  on an induced draft  cordwood
 furnace/water heaters.    Results were varied,   but low emissions
were  obtained   under   proper conditions  (estimated  from  carbon
monoxide and carbon  dioxide concentrations in flue gases).   High
temperatures  were  achieved by  a refractory  lined furnace  and a
 refractory secondary combustion chamber.
     When  a  fuel   charge  is  added  to a firebox at   very high
temperatures,   the  wood volatilizes very  rapidly.  While  this  can

                                                            PAGE 46

-------
                                      WOODSTOVE DESIGN FACTORS

 provide   very efficient combustion,  the critical requirement  is
 supplying adequate air to match the evolving volatile  compounds.
 Very  high  temperatures are possible with this type  of  system.
 The   high rate of heat output from such systems require some type
 of heat  storage or moderation.

   D.  Microprocessor Control
      The  batch process of combustion in cordwood stoves  presents
 a great challenge.   Changing temperatures,  fuel gas  concentra-
 tions  and  combustion  air requirements make  designing  an  air
 supply  system which will provide the proper amount of air  under
 all   conditions difficult.   Even mechanically drafted stoves can
 create poor combustion conditions,  primarily due to the  "batch"
 process of fueling.
     Some  design  work is underway to apply microprocessor  con-
 trols  to wood stoves.  System monitoring of stack oxygen  levels
 or  firebox  temperatures  could provide feedback  to  reduce  or
 increase the combustion air supply, compensating for rich or lean
 conditions.   For cordwood stoves not utilizing catalytic combus-
 tors, this approach may be necessary to achieve very low particu-
 late emission standards.  The high cost of developing and instru-
menting  such  a system appear to be the factors  preventing  the
 application  of  existing boiler technology to  residential  wood
combustors.   Few if any small woodburning appliances with micro-
processor controls are currently available.
                                                           PAGE 47

-------
                                      WOODSTOVE DESIGN FACTORS


 References
 1.   Marshall,  N. ,  Patten/  B.,  and Skog,  K.,  "The Dynamics of
 Residential  Wood-Energy  Use  in the United  States:  1970-2030 r"
 Resource Policy Center, Dartmouth College, 1983.

 2.   Skog,  K. and Watterson, I., "Residential Fuelwood Use  in the
 United States:  1980-81,"  U.S. Department of Agriculture,  Forest
 Service, July 1983.
                                                                 "
3.   Thompson,   C.,   "Residential  Energy  Consumption  Survey:
Consumption  and  Expenditures  April 1982  through  March  1983
DOE/EIA - 0321/2 (82).

4.  Wood  'N1 Energy,  "Marketing in the '80's",  Special Seminar
Edition, 1984.

5.  Federal Register,  Vol.  50. No. 149, Friday, August 2, 1985,
"Standards of Performance for New Stationary Sources; Residential
Wood Combustion", p. 31504.

6.  Burnet,  P.,  Edmisten,  N., Tiegs, P., and Yoder, R., "Acid-
Potential  Emission Factors for Residential Wood Buring  Stoves,"
Air Pollution Control Association, Annual Meeting, June 1985.

7.  Burnet,  P. and Tiegs, P., "Woodstove Emissions as a Function
of Firebox Size",  Paper 84-27,  21st Annual Meeting, PNWIS/APCA,
November 12-14, 1984.

8.  "Standard Test Methods for Heating Performance and Emissions
of  Residential  Wood-Fired  Closed  Combustion-Chamber   Heating
Appliances",   American   Society   of  Testing  and   Materials,
September, 1985.

9.  Fuentes,  K.  and Hodas,  L.,  "Feasibility Study of Enhanced
Combustion  Via  Improved  Wood  Stove  Firebox  Design,"  Radian
Corporation, Wood Heating Alliance, Baltimore, March 16-19, 1985.
ACKNOWLEDGEMENTS


     The  contributions and assistance of Professor Philip Malte,

Department of Mechanical Engineering,  University of  Washington,

are greatly appreciated.

-------
          APPENDICES





A.  STOVE DESCRIPTIONS




B.  STOVE TEST DATA




C.  GRAPHICAL DATA PRESENTATION

-------
                                      WOODSTOVE DESIGN FACTORS
.Notes on Stove Codes

Stoves are identified by an alpha-numeric code:

  *  The  first alpha character(s)  (A through EE) designates  the
     stove manufacturer and model;

  *  The first numeric character, immediately following the alpha
     code but proceeding the hyphen,  designates the test  series
     by  a manufacturer at OMNI labs.   Unless otherwise noted in
     Appendix  A, Stove Descriptions (see  Comments  column),  no
     modification was made to the stove between test series.

  *  The  number  following the hyphen represents  the  test  run
     number in the manufacturer's testing series.

Example:
     A2-7
     ttl
               test run number in second test series on stove A
            second test series on stove A
      	 Stove manufacturer/model code
                                     »
Appendix  A,  Stove  Descriptions,  shows  modifications  to  the
stoves.

Appendix B presents stove test data identified by stove codes.

Appendix  C  shows  stove test data from Appendix  B  plotted  in
graphical form.  Data points are identified by stove code.

Additional  stove  test data is compiled in a separate  companion
volume entitled "Data Supplement - The Effects of Design  Factors
on  the Emissions from Non-Catalytic Residential Wood  Combustion
(RWC)  Appliances."

-------
                                       WOODSTOVE  DESIGN  FACTORS
     The  following   information   represents  compiled  data  from
woodstove   emissions   and  performance   testing   conducted   in
accordance  with  the Oregon Department of  Environmental  Quality
"Standard Method for Measuring the Emissions  and Efficiencies  of
Residential Woodstoves," June 8f 1984.
     Descriptions  of   the stove technologies  include  pertinent
information on the size and configuration of  stoves referenced in
the data.  This is followed by test data grouped by stove model.

-------
    APPENDIX A




STOVE DESCRIPTIONS

-------
Stove
 Code

Al

A2
Firebox
 (FT3)

  1.6

  1.2
Bl
B2
Cl
C2
Dl
D2
2.1
2.3
2.3
1.4
1.8
1.6
            2.6
Dimensions (in.)   Primary
 WHO       Air

19.5  13.9 10.4    a,b,c

19.5  10.6 10.4    a,b,c
                    16.8  12.5 17.0    a,e,f

                    16.9  13.0 18.5    a,e,f


                    17.0  12.3 19.1    a,b,c

                    31.1  11.1 17.0    a,b,c


                    16.8  10.0 19.0    a,b,c


                    14.7  10.3 17.9    a,b,c
          22.9   20.0   9.8    e,f,h,i
Secondary
  Air

   no

   no
                                         no
                                         no
                               c,f
Refractory
 Lining

  yes

  yes
Firebox
 Breach

   g

   g
no
no
yes
no
no
see comments
g
g
g
g
                                          yes
                                          see
                                          comments
no
                     Comments
Actual depth 17"
- log irons
prevent use of
forward 6.5". A2
had   grate to
reduce fuel load
with no change
in actual firebox
dimensions. Con-
vection heater.

Firebrick removed
for B2.
Convective
heater.

Cast iron fire-
box liner used
to reduce fire-
box size on C2.

Dl - brick on
floor only.

D2 - Fully
lined w/fire-
brick & reflec-
tive plates.

Stove and fire-
box made of cast
iron. Primary
air controlled
by a thermostatJ

-------
Stove
 Code

F.l

F2
G2
HI

H2
Firebox   Dimensions (in.)   Primary
 (FT3)      WHO       Air

  1.3     18.8  12.0 10.0    c,d,o

  1.44     19.5  12.1 10.5    c,d,o
  2.04    20.6  15.4 11.5    c,d,o
  1.0     17.0  10.5  9.5    b,e,p

  1.0     17.0  10.5  9.5    b
            2.2     24.0  13.0 12.0    e,p
            1.7     11.5  12.0 21.0    c,f,i
            2.1     13.9  17.7 14.0    a,e,f
Secondary
  Air
   no

   no
   no
   a

   a
                                         no
                                         no
                                         f,r
Refractory
 Lining
  no

  no
                                                               no
  no

  no
              no
              see
              comments
Firebox
 Breach

   m

   m
                            m
   g

   g
                m
Comments
Back baffle and
front grate
were in place.
Stove & fire-
box are made of
cast iron.  Fl
and F2 are the
same stoves.
The firebox was
remeasured for
F2 to reflect
reevaluation of
complex geometry.

Top baffle &
grate were in
place. Stove is
made entirely of
cast iron.

Cast iron fire-
box. Air on H2
entered along
sides of firebox
floor. Steeply
sloping baffle.

Underfire air
used in combus-
tion area. Cast
iron firebox.

Stove  is  of
down draft
design.

Refractory
lining on the
bottom, sides
and in chamber
below firebox.•
Downdraft desju

-------
Stove     Firebox   Dimensions  (in.)  Primary   Secondary   Refractory   Firebox   Comments
 Code       (FT3)     WHO       Air       Air        Lining       Breach

!•           0.8     11.0   8.5  15.0    b,c,o       f          no          see      Stove  is made
                                                                         comments  of cast iron.
                                                                                   Secondary air
                                                                                   directed into
                                                                                   narrow breach
                                                                                   at rear of
                                                                                   firebox.

M            1.3      16.5  10.0 13.8    a,c,f,i     c,f,t      yes           g    M2 runs  were
                                                                                   conducted in
                                                                                   efforts to
                                                                                   improve
                                                                                   "Oregon
                                                                                   weighted"
                                                                                   emission
                                                                                   values.

N    •       1.7     20.0  15.9  9.3    b,e,i       no         v             q      Cast iron fire-
                                                                                   box surrounded
                                                                                   by a stainless
                                                                                   steel cabinet.

0           2.5     18.5  14.0  16.8    a,e,f       no         yes           u      Stove uses con-
                                                                                   vection air
                                                                                   blower.

P           1.6     .16.0  17.5  10.0    c,p         f          no            g      Convection
                                                                                   heater.

Ql          1.1     22.0   9.5  9.0    a,c,f       no         no            g      Ql & Q2 are
                                                                                   same stove with
Q2          1.3     22.0  11.5  9.0    a,c,f       no         no            g      different
                                                                                   interpretation
                                                                                   of firebox
                                                                                   volume changing
                                                                                   fuel load size.

-------
 Stove
 Code

 R
T


U
V
Firebox   Dimensions (in.)   Primary   Secondary
 (FT3)      W     H    D       Air       Air
  1.4
17.4   9.8 14.5    a,b,c,i
   1.3



   1.3

   1.2*
 16.6  12.0 12.0



18.3   9.5 12.8

11.5  13.0 14.0
 a,c,f



a,b,c

aa,bb,cc
        *fuel  placed  within combustion zone
           approx. 3" x 3"
  5.1
23.5  21.6 17.3
bb,cc
                               c,f ,x
                     Refractory
                      Lining

                       yes
                                                     Firebox
                                                      Breach
                                no



                               no

                               f ,cc
                                                               yes
yes

side &
rear walls
9

1
                               no
                                                              yes
                                                               Comments
                     A second second-
                     ary  air  tube
                     runs perpend-
                     icular to the
                     front tube down
                     the top center
                     of the  firebox.
                     Fireplace insert,

                     Firebrick on
                     floor and sides
                     of firebox.
Pellet stove
induced draft
Blower fan  &
fuel feed rate
are under sepa-
rate controls.
Fuel fed by an
auger & is
introduced from
underneath the
combustion zone.

Pellet fuel
add-on device on
large stove
using induced
draft system.
Blower fart and
fuel  feed  rates
are controlled by
separately. Fuel
is fed by an
augor and is
dropped into
the combustion
zone from above:

-------
Stove
 Code

W
Firebox   Dimensions (in.)   Primary   Secondary   Refractory
 (PT3)      W     H    D       Air       Air        Lining
  2.4
16.0  15.0 17.0    bb,cc
                                                   no
                                                    yes
           Firebox
            Breach

              u
          see comments
                             b,bb,cc
                                                              yes
            2.4
          17.0   13.5  18.0
                   aa,o,b
                                                   no
yes
                                                               Comments
Same pellet
add-on device as
V on smaller
stove.  Primary
air closed for
use with add-on
device. Induced
draft system.
Blower fan and
fuel feed rates
controlled sepa-
rately.

Pellet stove w/
a forced draft
exhaust system.
Blower fan and
fuel feed rates
are controlled
separately.
Fuel is fed by an
auger and is
dropped into the
combustion zone
from above.  New
test data; des-
criptions & data
being compiled.

Baseline stove/
very small
baffle.  Z2 runs
were conducted
with densified
fuel.

-------
Stove
 Code

AA
Firebox   Dimensions (in.)   Primary
 (FT3)      WHO       Air
  3.5
BB
  0.8
CC
  1.4
DD
  2.0
                            Secondary
                              Air
19.0  19.0 17.0    aa,b
                                                   no
 9.0  17.0  9.0    b,o,aa
                                                   no
18.0   8.0 16.75   a,c,f
no
18.0  14.0 14.0    a,c,f
no
         Refractory
          Lining

           yes
           Firebox
            Breach
                                                              no
yes
bottom,
rear/ &
3" on sides,
sildes

yes
bottom,
rear, &
sides
Comments
Baseline stove.
AA1 and AA2 runs
reflect no change
in stove design.
AA2 runs with
asterisk denote
ASTM fuel loads.

Top baffle
extending 2/3
the depth of
firebox from
the backwall
toward the
loading door.
Cast iron fire-
box.  BB1 runs
with asterisk
were conducted
with ASTM fuel
loads.  Emission
testing was not
conducted on runs
BB1-2 and BBl-4.

Uses brick
baffle plate.
By-pass for
"secondary
combustion
chamber".

-------
Stove
 Code

EE
Firebox   Dimensions (in.)   Primary
 (FT3)      WHO       Air
  2.2
FF
  2.4
15.0 diameter x    b,c,o
                            Secondary
                              Air

                               no
                                        Refractory
                                         Lining
Firebox
 Breach
                                                              no
23.0  15.0 12.0    e,h
                                                   yes
                                          yes
Comments
          Cylindrical
          firebox.
          Double paned
          window in front,
          0.5" x 1.5"
          deflectors at
          air inlets.
          Primary air
          enters through
          both sides.
          Right side
          primary air
          goes through
          loading door.
          Top baffle.

          Air enters
          firebox at
          rear bottom
          center.

-------
                                      WOODSTOVE DESIGN FACTORS

 STOVE  TECHNICAL DESCRIPTION
 Notes:
 a - Air  enters firebox through slot located at top front of
    firebox, directly over door.
 b - Little or no preheating of air.
 c - Air  slide control.
 d - Air  stop maintains high minimum burn rate.
 e - Thermostatic control.
 f - Preheated.
 g - Front top 1/3 of firebox.
 h - Air  enters firebox through openings located on the rear of
    the  firebox.
 i - Air  enters firebox through openings located at the bottom
    front of the fireplace.
 j  - Exits firebox at the lower right back corner.
 k  - Located in the retro fit device.
 1  - Exits at the top rear of the firebox.
 m - Exits at the upper center of the back wall.
 n  - No independent secondary air introduced; primary air used.
 o  - Air  enters firebox through an opening in the loading door.
 p  - Under fire air.
 g  - Exits the firebox at the bottom of the rear wall.
 r  - Secondary air flow unregulated.
 s  - Down through the bottom center, below the firebox floor.
 t  - Enters firebox horizontally across the top of the firebox.
 u  - Exits firebox at the top center of the firebox.
v  - Refractory secondary combustion chamber.
X  - Enters horizontally across the top front 1/4 to 1/3 of the
    firebox.

-------
                                      WOODSTOVE DESIGN FACTORS
aa - Air controlled by  spin draft.

bb - Air enters firebox through holes in burner rings surrounding
     a combustion zone.

cc - Air controlled by  a blower fan.

dd - "Door wash" air inlet (slot across top part of firebox, air
     directed torward glass).

ee - Air enters firebox  at both bottom front corners.

-------
   APPENDIX B





STOVE TEST DATA

-------
STOVE CODE

Al-2
Al-3
Al-4
A2-5
A2-6
A2-7
Bl-1
B2-1
B2-2
B2-3
Cl-1
Cl-2
Cl-3
Cl-4
C2-1
C2-4
C2-5
C2-6
C2-2
C2-3
01-1
01-2
02-2
02-3
02-4
D2-6
El-1
El-2
El-4
El-34
Fl-1
Fl-2
F2-1
F2-2
F2-3
F2-4
F2-6
F2-7
G2-B
G2-9
G2-20
C2-:i
Hl-3
Hl-4
FIREBOX BURN RATE BEAT OUTPUT P ARTICULATE PARTICULATE PARTICULATE CO CO CO CONDUSTION HEAT OVERALL
SHE (LBS/UR) (BTU/HR) EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EFFICIENCY TRANSFER EFFICIENCY
("3)
1.6
1.6
1.6
1.2
1.2
1.2
2.1
2.3
2.3
2.3
2.3
2.3
2.3
2.3
1.4
1.4
1.4
!..«
1.4
1.4
1.8
1.8
1.6
1.6
1.6
1.6
2.5
2.5
2.S
2.5
1.3
1.3
1.4
1.4
1.4
1.4
1.4
1.4
2.0
2.0
2..0
2.0
1.0
1.0

5.2
9.5
7.7
5.3
8.8
12.0
4.6
3.5
4.3
8.5
2.1
5.4
6.1
3.5
2.4
1.3
7.9
3.6
2.6
3.5
3.1
2.7
2.0
2.8
3.9
6.4
7.6
5.6
3.7
1.6
4.9
3.3
5.0
10.3 .
7.3
8.4
6.4
7.3
4.9
7.6
12.6
6.6
5.9
4.7
(G/KG) (G/UR) (G/106 JOULE)
24411.0
35126.0
30621.0
2S054.0
37257.0
45450.0
21267.0
16874.0
21074.0
42743.0
8780.0
23933.0
26780.0
15438.0
11760.0
5958.0
35473.0
16198.0
12766.0
17297.0
13383.0
11219.0
8966.0
13499.0
18851.0
30062.0
31658.0
22211.0
14181.0
6545.0
22947.0
15529.0
25214.0
44336.0
26281.0
38251.0
29707.0
33880.0
24203.0
36547.0
54498.0
30582.0
25999.0
21362.0
5.4
1.6
3.8
2.2
2.3
1.2
6.2
11.2
7.9
1.6
44.6
6.5
5.7
14.2
9.3
36.3
4.3
14.0
10.3
2.6
15.8
21.5
20.9
18.4
4.2
2.2
15.9
33.3
31.3
31.7
7.3
15.9
4.2
2.2
2.5
1.3
11.1
7.0
6.9
8.9
5.3
3.1
12.3
16.4
10.3
5.8
10.9
4.4
7.5
5.3
10.7
14.6
12.8
4.9
34.6
13.2
13.0
18.3
8.3
17.6
12.8
18.3
9.9
3.3
18.2
21.0
15.5
19.3
6.1
5.3
44.9
69.3
43.0
18.9
13.2
19.3
7.6
8.6
6.8
4.0
26.6
18.7
12.6
25.1
24.6
7.6
26.6
28.3
0.4
0.2
0.3
0.2
0.2
0.1
0.5
0.8
0.6
0.1
3.7
0.5
0.5
1.1
0.7
2.8
0.3
1.1
0.7
0.2
1.3
1.8
1.6
1.4
0.3
0.2
1.4
3.0
2.9
2.7
0.5
1.2
0.3
0.2
0.2
0.1
0.8
0.5
0.5
0.7
0.4
0.2
1.0
1.3
(G/KG)
128.0
38.5
82.8
104.6
51.8
2?. 5
117.5
166.6
122.6
64.2
327.7
120.2
118.3
143.2
183.1
328.0
73.7
163.1
167.8
124.3
180.4
175.8
278.8
190.5
117.2
90.4
144.6
187.3
248.5
426.2
116.7
.165.2
73.6
27.0
50.6
38.7
68.6
57.3
167.4
145.1
104.0
120.5
149.5
151.1
(G/HR) (G/106 JOULE)
245.0
135.1
235.0
209.3
168.9
104.9
202.1
217.4
199.2
202.8
254.4
242.5
267.4
184. 7
163.6
159.2
216.6
212.9
161.1
159.8
207.6
171.6
206.2
200.8
169.0
215.7
407.0
389.0
342.6
253.5
210.6
201.5
134.8
105. 5
135.7
120.2
165.2
154.4
304.9
409.0
484.2
293.4
323.7
261.4
9.5
3.6
7.2
7.9
4.3
2.2
9.0
12.2
9.0
4.5
27.5
9.6
9.5
11.3
13.2
25.3
5.8
12.5
12.0
8.8
14.7
14.5
21.8
14.1
8.5
6.8
12.6
16.6
22.9
36.7
8.7
12.3
5.1
2.3
4.9
3.0
5.3
4.3
12.0
11.0
B.6
3.7
11.8
11.6
EFFICIENCY
84.1
87.0
85.3
89.1
86.4
88.4
84.4
82.7
84.1
87.3
74.1
83.6
84.4
81.7
80.6
74.1
85.3
81.1
82.3
84.1
79.0
79.8
75.1
80.2
84.4
84.0
83.9
77.5
74.7
77.7
87.0
82.0
87.8
89.1
87.1
89.1
83.5
83.1
81.7
82.1
83.6
84.0
B1.9
81.7
78.7
59.6
66.0
76.5
68.7
60.3
75.9
81.2
80.0
80.2
77.8
72.9
72.4
75.8
84.7
85.2
73.0
79.3
83.8
83.0
79.6
77.6
82.2
81.7
79.5
77.1
69.9
72.4
73.4
75.7
75.6
80.6
80.4
66.1
58.3
70.6
76.6
78.3
84.0
80.1
71.5
77.5
76.3
78.5
66.1
51.9
56.3
65.7
59.4
53.3
64.1
67.1
67.3
70.1
57.6
61.0
61.1
61.9
68.2
63.1
62.2
64.3
68.9
69.8
62.9
62.0
61.7
65.6
67.1
64.8
58.6
56.1
54.8
58.8
65.6
66.1
70.6
58.9
50.8
62.9
64.0
65.0
68.6
65.7
59.8
65.1
62.5
64.1

-------
 Hl-1
 H2-6
 H2-8
 H2-9
H2-10
H2-11

 11-1

 Jl-1

 Kl-1
 Kl-2
 Kl-3
 Kl-4

 Ll-1
 Ll-2
 Ll-J
 Ll-4
 Ll-5
Ll-19
 Ml-1
. Ml-2
 Ml-5
 Hl-6
 Ml-7

 M2-1
 M2-2

 N2-1
 N2-2
 N2-3
 N2-4
 N2-5
 N2-6
 N2-7

01-1B
01-28
01-3B
01-4B

 Pl-1
 Pl-2
 Pl-3

 Ql-2
 01-3
            SI SB
           (PT3)
1.0
1.0
1.0
1.0
1.0
1.0

2.2

1.7

2.1
2.1
2.1
2.1

0.8
0.8
0.8
0.8
0.8
0.8
1.3
1.3
1.3
1.3
1.3

1.3
1.3

1.7
1.7
1.7
1.7
1.7
1.7
1.7

2.5
2.S
2.S
2.5
1.1
1.1
ORN RATE HEAT OUTPUT PARTICIPATE PARTICOLATE PART1CULATE
(LBS/HR) (BTU/HR) EMISSIONS EMISSIONS EMISSIONS
(G/KG) (G/HR) (G/106 JOULE)
4.1
5.1
4.9
5.1
6.0
6.9
3.0
3.6
2.7
5.1
3.2
5.5
3.0
2.8
1.9
6.9
4.1
2.0
6.2
2.2
9.5
3.0
4.8
7.8
2.9
13.7
9.0
7.0
6.7
11.1
B.I
8.5
9.6
5.5
4.6
10.4
10.9
5.1
4.7
4.1
14.4
16303.0
23749.0
21721.0
23582.0
20660.0
24934.0
12611.0
18557.0
12546.0
24280.0
15778.0
27163.0
12806.0
11800.0
6887.0
32043.0
18279.0
8209.0
22118.0
8961.0
29181.0
14510.0
20116.0
35198.0
12945.0
57260.0
39260.0
' 31203.0
30311.0
48973.0
36049.0
36784.0
43775.0
27609.0
22234.0
44960.0
34S09.0
20726.0
21641.0
20376.0
60100.0
30.7
16.3
6.4
8.4
5.7
6.9
44.9
6.3
19.7
13.1
16.7
3.6
17.1
24.4
22.0
1.4
4.5
3.9
12.7
24.2
4.8
7.0
6.9
6.4
14.5
2.7
3.2
4.3
5.4
1.8
5.5
4.2
3.5
2.7
22.6
3.5
8.2
6.7
10.9
6.1
4.2
46.1
30.4
11.4
16.0
12.7
17.9
50.4
8.5
20.0
25.1
19.9
7.4
20.3
24.9
36.3
3.5
7.7
2.6
28.8
19.7
16.9
7.6
12.2
18.8
15.9
13.4
10.7
11.3
13.4
7.4
16.3
13.2
12.7
5.6
38.4
13.5
33.5
12.6
19.2
9.5
22.2
2.7
1.2
0.5
0.6
0.6
0.7
3.8
0.4
1.5
1.0
1.2
0.3
1.5
2.0
2.1
0.1
0.4
0.3
1.2
2.1
0.6
0.5
0.6
0.5
1.2
0.2
0.3
0.3
0.4
0.1
0.4
0.3
0.3
0.2
1.6
0.4
0.9
0.6
O.B
0.5
0.
-------
 01-5
 Q2-6
 02-7

 Rl-1
. Rl-2
 Rl-4
 Rl-5
 Rl-6
 Rl-7

 Sl-1
 Sl-2
 Sl-3
 Sl-4
 Sl-5

 Tl-1
 Tl-2
 Tl-3
 Tl-4
 Tl-5
 Tl-6
 Tl-7
Tl-10
Tl-11
Tl-12
Tl-13
Tl-14

 Ul-1
 Ul-2
 Ul-3
 Ul-4

 Vl-1
 Vl-2
 Vl-3
 Vl-4

 H2-5
 W2-6
 H2-7
 W2-8

 Xl-1
 Xl-2
 Xl-3
 Xl-4
11-20
Zl-21
Zl-22
Zl-23
EBOX BURN RATE BEAT OUTPUT PARTICOLATE PARTICULATE PARTICULATE CO CO CO COMBUSTION
HE (LBS/HR) . (BTU/HR) EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EFFICIENCY
T3)
1.1
1.3
1.3
.4
.4
.4
.4
.4
.4
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.1
1.1
1.1
1.1
S.I
S.I
5.1
5.1
2.4
2.4
2.4
2.4
0.3
0.3
0.3
0.3
2.4
2.4
2.4
2.4

S.2
5.2
4.7
6.1
3.0
2.5
2.3
3.4
7.4
2.5
5.2
9.9
1.4
4.1
4.2
8.2
3.7
2.7
2.0
3.4
5.1
11.1
2.6
1.3
4.3
3.0
4.4
1.6
2.6
3.3
1.4
1.8
3.0
9.3
1.2
2.0
9.2
3.0
1.5
2.3
3.0
4.0
4.1
2.3
2.8
11.7

25012.0
24346.0
20840.0
26905.0
14721.0
12065.0
10543.0
15864.0
30258.0
12447.0
23096.0
40993.0
6259.0
18500.0
17890.0
28045.0
17234.0
10455.0
8256.0
13311.0
23045.0
43350.0
13626.0
4887.0
18154.0
14717.0
29192.0
8744.0
17945.0
22745.0
7903.0
10866.0
18345.0
55349.0
6932.0
12308.0
56196.0
17961.0
9499.0
15041.0
20126.0
26638.0
18926.0
10544.0
10243.0
47569.0
(G/KG)
2.8
2.3
2.9
9.3
9.3
11.4
11.4
6.2
4.3
12.9
8.1
5.0
19.3
11.2
9.5
9.8
19.2
28.4
30.1
20.4
9.9
4.9
11.8
34.9
9.2
7.4
0.7
1.0
0.6
0.6
4.0
1.8
2.2
0.8
6.0
3.1
1.9
1.8
0.9
1.0
1.0
1.4
19.8
28.7
24.3
9.9
(G/HR) (G/106 JOULE)
5.3
4.5
5.0
21.0
10.5
10.7
9.6
7.8
11.8
12.1
15.9
18.4
10.0
16.9
14.9
30.4
26.3
28.6
21.9
26.0
19.2
20.0
12.4
16.7
14.6
8.4
1.2
0.7
0.7
0.9
2.3
1.4
2.8
3.2
2.9
2.6
7.3
2.2
0.5
0.9
1.2
2.4
29.7
24.1
25.3
42.7
0.2
0.2
0.2
0.7
0.7
0.8
0.9
0.5
0.4
0.9
0.7
0.4
1.5
0.9
0.8
1.0
1.4
2.6
2.5
1.9
' 0.8
0.4
0.9
3.2
0.8
O.S
0.0
0.1
0.0
0.0
0.3
0.1
0.1
0.1
0.4
0.2
0.1
0.1
0.1
0.1
0.1
0.1
1.5
2.2
2.3
0.9
(G/KG)
97.9
111.7
127.9
128.1
150.7
179.7
212.2
150.7
78.5
194.1
145.6
94.8
237.2
150.1
103.8
134.5
133.8
245.1
310.1
157.1
95.8
56.4
201.7
377.4
179.0
175.0
23.6
29.1
5.2
1.9
34.7
17.7
17.5
4.3
32.6
18.3
44.3
8.0
29.1
37.5
17.8
22.1
172.5
297.1
238.3
114.6
(G/HR) (G/106 JOULE)
187.3
218.3
224.2
289.3
170.6
167.9
178.0
189.0
214.3
181.7
286.5
347.7
122.4
226.0
162.3
417.2
183.6
247.1
225.6
199.4
184.8
228.7
212.8
181.0
285.4
200.3
43.1
19.3
5.7
2.4
19.8
13.5
22.7
17.0
15.9
15.6
171.8
10.0
17J4
34.8
22.0
36.5
259.0
249.2
248. 5
491.8
7.1
8.5
10.2
10.2
11.0
13.2
15.9
11.3
6.7
13.8
11.8
8.0
18.5
11.6
8.6
14.1
10.1
22.4
25.9
14.2
7.6
5.0
14.8
35.1
14.9
12.9
1.4
2.1
0.3
0.1
2.4
1.2
1.2
0.3
2.2
1.2
2.9
0.5
1.7
2.2
1.0
1.3
13.0
22.4
23.0
9.8
BEAT OVERALL
TRANSFER EFFICIENCY
EFFICIENCY
88.8
64.7
82.9
80.8
82.8
80.4
79.9
82.0
82.7
80.6
78.1
84.4
76.1
80.8
87.6
92.5
82.1
72.8
74.1
82.2
84.5
90.2
84.2
69.2
62.9
63.3
92.6
91.9
99.0
94.3
87.7
88.9
90.5
92.3
91.8
93.8
90.1
91.7
89.0
69.1
91.0
92.3
81.7
77.8
77.5
84.6
76.2
75.2
74.5
76.5
80.7
82.9
61.8
80.4
69.2
84.2
77.6
69.3
65.6
78.8
70.2
51.9
81.6
75.7
81.3
67.5
75.1
63.1
81.6
80.8
73.4
78.5
87.2
74.7
84.9
87.6
81.6
83.0
81.3
78.9
80.3
60.2
63.4
61.2
87.9
89.6
87.7
86.2
81.9
85.3
85.3
72.5
67.7
63.7
61.8
61.8
66.8
66.6
65.4
65.9
57.2
67.9
60.5
58.5
65.3
63.7
61.6
48.0
67.0
55.1
60.2
55.5
63.5
56.9
68.7
55.9
60.8
65.3
80.8
68.7
84.0
62.7
71.6
73.8
73.6
72.8
73.7
75.3
75.1
74.5
78.2
79.8
80.1
79.5
66.9
66.4
65.8
59.2

-------
   12-1
   12-2

  AA1-6
  AA1-7
  AA1-8
  AA1-9.

•  AA2-8*
  AA2-9*
  AA2-10
  AA2-11
 AA2-12*
 AA2-13*

  BB1-1*
  BB1-2*
  BB1-4*
  BB1-5* ,
  BB1-6*
  BB1-7*

  CC1-15
  CC1-14
  CC1-16

  DD1-9
  DD1-36
  DD1-35
  DD1-6

 EE1-1**

  PF1-18
  FF1-21
  PP1-19
  FF1-17
EBOX BORN RATB BEAT OUTPUT PARTICULATE PARTICOLATB PARTICULATE CO CO CO COMBUSTION HEAT OVERALL
lit (LBS/BR) (BTO/HR) EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EMISSIONS EFFICIENCY TRANSFER EFFICIENCY
1*3) (G/KG) (G/HR) (G/106 JOOLE) (G/KG) (G/HR) (G/106 JOOLE) EFFICIENCY
2.4
2.4
3.5
3.5
3.5
3.5
3.5
3.5
3.5
3.5
3.5
3.5
o.a
o.a
0.8
0.8
0.8
o.a
1.4
1.4
1.4
2.0
2.0
2.0
2.0
2.2
2.4
2.4
2.4
2.4
4.0
2.2
4.6
2.2
3.0
7.6
6.4
5.9
2.1
2.6
B. 3
8.6
6.2
6.1.
2.1
1.7
11.0
10.5
3.7
5.8
10.2
2.0
3.0
4.3
14. »
4.6
2.0
2.8
3.9
6.6
24728.0
13432.0
20662.0
9159.0
11980.0
31615.0
27050.0
25781.0
8596.0
10934.0
36266.0
35213.0
29670.0
26301.0
9834.0
7919.0
44600.0
46090.0
18590.0
25342.0
36058.0
7919.0
13331.0
19606.0
62447.0
20484.0
9272.0
11115.0
17868.0
30123.0
9.4
. 12.8
22.4
28.2
29.3
7.2
13.7
22.3
24.7
32.0
10.6
12.2
9.2
-
-
33.2
3.4
2.3
12.5
11.7
2.5
52.5
27.7
17.5
8.5
31.6
28.8
37.8
24.3
9.2
15.9
11.8
39.2
23.2
32.9
20.0
32.1
48.4
19.0
30.9
32.3
38.9
21.6
-
-
21.1
13.5
6.9
17.3
24.7
9.3
38.8
31.1
27.6
47.7
53.4
22.1
38.2
35.6
22.6
0.6
0.8
1.8
2.4
2.6
0.6
1.1
1.8
2.1
2.7
0.8
1.0
0.7
-
-
2.5
0.3
0.2
0.9
0.9
0.2
4.6
2.2
1.3
0.7
2.5
2.3
3.3
1.9
0.7
73.4
97.7
215.7
298.2
281.4
118.4
169.2
206.2
362.2
328.2
139.6
162.2
154.9
168.4
186.6
250.4
16.0
29.6
119.8
100.7
44.4
352.2
211.4
127.4
104.1
226.5
321.9
305.9
199.4
120.4
124.9
90.0
374.9
247.4
309.9
336.9
398.2
446.9
279.2
316.1
423.2
518.5
364.3
377.9
144.7
159.3
72.9
117.3
164.7
211.1
167.3
259.8
237.6
200.6
579.6
382.5
246.4
308.3
292.1
295.5
4.8
6.4
17.2
25.6
24.5
10.1
14.0
16.4
30.8
27.4
11.1
14.0
11.6
13.6
13.9
19.1
1.5
2.4
8.4
7.9
4.4
31.1
16.9
9.7
8.8
17.7
25.2
26.3
15.5
9.3
89.5
89.1
80.6
73.1
75.7
85.5
80.5
76.9
70.7
71.7
82.3
79.0
80.1
78.5
79.0
76.0
91.6
88.2
83.9
87.1
89.8
70.0
77.0
83.9
64.7
77.8
75.9
75.4
81.5
88.1
81.0
81.7
78.1
85.3
79.8
72.7
73.7
79.8
82.0
81.3
75.2
75.8
80.4
• 76.6
83.6
84.5
62.5
68.6
81.7
70.3
57.0
79.3
83.7
75.6
69.2
80.8
82.1
77.4
77.4
74.3
72.5
72.8
63.1
59.8
56.9
60.0
59.4
61.4
58.0
58.3
61.9
59.9
64.4
60.1
66.0
64.2
57.2
60.5
68.6
61.3
51.1
55.5
64.5
63.4
58.6
62.8
62.3
58.4
63.0
65.4
  •Fuel load  based on proposed  ASTM procedures  (Reference 8).  Fuel
      loading density was approximately 12 to 13  pounds fuel per usable
      cubic foot of firebox volume.
  **Partlculate samples obtained with Condar sampler.

-------
          APPENDIX C





GRAPHICAL DATA PRESENTATION

-------
FIREBOX  SIZE  EFFECTS


1?
o

-------
                              BASELINE   STOVES
       70 -i
oc
__/
o
X
if
ir
o
i/>
f/)
^£
C
LJ

§
->
U
CL
60 -
       40 -
.30  -
20  -
       10 -
        0
1 -20
           0
                    Baseline stove data is used as a refer-
                    ence to approximate emissions from
                    "typical" stoves found In many Install-
                    ations. This stove had a 2.4 ft3 fire-
                    box with a refractory brick lining, a
                    splndraft control bringing air in
                    through the lower 1/3 of the door. It
                    had a very small baffle directly below
                    the flue.
                                                          21 -2,3
                        20                   40
                                   f Thou sands')
                            H EAT  O Cl TP LI T  (BTlf/ HO U Rj
                                     Figure C-2
                        Firebox Modifications to Stove Z
                                          60

-------
40 -,
tt?' 35 -
r>
I
to" w
3 25 -
Q 20-
f.O
1.0
^ 15 -
LJ
b
B 10-
F;
£ 5 -
0 -i
c
'FIREBOX S
ZE EFFECTS
The firebox volume of 2.3 ft3 in Cl tests was
reduced to. 1.4 ft3 in C2 tests with the
addition of refractory bricks and a cast iron
C1 — 1 freestanding firebox liner. This reduced the
fuel load from about 16 Ibs to just over 9 Ibs.
Significant emissions reductions are seen at the
lower heat output levels, although some overlap
of performance is seen.

/••* o .1 \-'-i f- •VD
Oz— 4-


01-62-3 C2-5
02-1

i i i
D 20


i i
40 60
           (Thousands)
    HEAT OUTPUT (BTU/HOUR)
            Figure C-3

Firebox Modifications to Stove C

-------
                             Rl i -  »•-•*• f • •- »"•". '.  X   j**™»» I •""• -y f ' ••   -—™» ••»•» «*MB r	 .•—. — — „••„
                             REBOX   SIZE   EFFECTS
o
ft:
        40



        35


        30
       25
                                       01 tests were conducted on a 1.8 ft^ steel
                                       firebox with about 12 Ibs of fuel. D2 testa
                                       were conducted on 10.3 Ibs of fuel in a
                                       1.6 ft3 firebox with refractory brick and
                                       reflective metal in place.  This stove
                                       appears to be relatively sensitive to burn
                                       rate changes, as emissions drop rapidly
                                       at medium high burns.
O
Co
 LJ

 LJ


.5
1
       20 -
       15 -
       10 -
         ft, _
                     01 -2
                  D2-2
                                         02-6
        0
0
                                20
n	r~
           40
 (Thousands)
 •	'ITU/
                                                                           60
                                   HEAT OUTPUT (BTU/HOUR)
                                            Figure  c-4

                                Firebox Modifications to Stove D

-------
                                   FLREBO
                                                O ri
                                 L
       -4-0 -r
  ..
tt:
_/
o
o:
CO
LJ
U
r.;

I
       .30 -
25 -J
        1 0 -
  5 -J
                                           F2-6
                         D2-5
                        02--4
                      Both D2 and F2 runs were conducted on
                      1.6 ft-* usable firebox stoves.  The
                      actual height of the F2 stove was
                      about 16 Inches, although It was
                      measured at about 10" due to the fuel
                      loading door configuration preventing
                      fuel being placed In the upper  fire-
                      box area.  The D2 stove had an  air
                      Intake slot over the door while the
                      F2 stove used an air slide control In
                      the lower section of the fuel loading
                      door. Fuel loads were about 1  Ib
                      lower on the F2 stove.
                                                     F2-4
            0
20
                                                                        60
                                              ('Thou sane? a;)
                                     MEAT  -OUTPUT  (6TU/HOUR)
                                              Figure C-5
                              Firebox/Fuel  Load Effects  on Emissions

-------
                             FIREBOX  HEIGHT
      70
                             C.2—
O
I
2
O
0

I
      60 -
      50 -
      40 -
      30  -
      20 -
       10 -
        0
                                           A range of firebox heights is shown below.
                                           Low firebox heights are usually associated
                                           with smaller firebox sizes. Firebox heights,
                                           in inches, are as followsi E • 20, G- IS,
                                           X • 18, L - 9, M - 10, 0 • 14.
                      E2-4
                                      E2-1
                             01-3B
                                     M1-3
                              M1-1
                    L1-2       K1-2      G2-9
                     M2-2          Ml-5
                            M1-92- 8
                      M1 -161 - 5
                LI -19
                             20                  40
                                       (Thousands;
                                HEAT OUTPUT (BTU/HOUR)

                                        Figure  C-6

                          Firebox  Height Effects  on Emissions
                                                                     r
                                                                    60

-------
                                       AIR   INLETS
       70
o:
t/
f/}
g
t/)
£'
2
LJ
c.;
i
60 H
 X
       50 -
       4-0 -
30 -
       20>
       10 -
         0
                                 tz—
Four stoves with similar sizes but different air inlet
designs are ooipared below.  Firebox sizes for stoves,
E and 0 were 2.5 ftj, while stoves B and C had 2.3 ftj
fireboxes.  Stove E had an air inlet very low in the
firebox, while stoves B, C, and 0 had air inlets through
slots above the fuel loading doors (door window wash
design).
                         EI-4-
                                           E2-1
                          01-30
            C1-1
                   032H6
                     02-1
                           H-1
                                       O1-2B
                                                         B2-3
            0
                         —1	1	1	1	T~
                          20                    40                   60
                                     (ThoLisonds)
                             HEAT  OUTPUT (BTLT/HOUR)
                                       Figure C-7
                           Emission  Reductions by Air  Inlet Location

-------
ND
   -
MB.
AMBER
                            \


t 	
tiL
->
O
^
i/;'
!5
(fr
&
1/3
o
LJ
I.|J
<•?
13
U
F
o:
•=!.
CL


t-u -

35 -



30 -
•J W


25 -

20 -
15 -


10 -




5 -
0 -
All data points shown were conducted under
the same stove configuration, with the
exception of Rl-1 (leveling bolt plugs
were not installed, allowing air leakage
through the firebox floor from a convec-
tion fan blower). This design utilizes
extensive secondary air Inlets, parallel
and perpendicular to the loading door.
Consistent performance at varying,
especially low, heat outputs, is unique
in this natural draft stove.
R1-1



R1-7
KR1&,V"2
w
R 1 - 6




I i i ii i
0 20 40 60
      (Thousands)
HEAT OUTPUT (BTU/HOUR)

       Figure C-8

Effective Secondary Air

-------
                                         DOWN DRAFT
n
x
•=r
i'c
CJ
o
TO
0
LJ
LJ
hp
1]
O
F
       JD -
       25 -
       20 -
        15 -
        10 -
         5
         0
                                         _ r>
                        K1-KH-3
                               J1-1
                                          K 1 - 4
            0
                                                        These two stove designs bring combustion
                                                        air down past fuel logs, into a burning
                                                        zone and DOWN away from the fuel before
                                                        being allowed to draft upwards.  Some
                                                        individual runs appear promising, others
                                                        show higher emissions.  Results from the
                                                        Kl stove may be misleading, however, as
                                                        an unusually large coal bed was necessary
                                                        to comply with test requirements. Normal
                                                        operation uses a smaller coal bed,
                                                        which does not block the firebox
                                                        breach to as great an extent.
                                                           T
                                   20                     40
                                               (Thousands)
                                       HEAT OUTPUT (BTU/HOUR)

                                                 Figure C-9
                                   Emissions  from Downdraft  Stoves
                                                                                  60

-------
                                         DOWN DRAFT
       70
'o
 I

 o
 (7;
 LJ

 UJ
.0
       60 -
       50  -
       30  -
       20 -
       10 -
         0
            K1-2
K1-K1-3

     B2-;
          12-2
          M-1
                      A ,"well designed" downdraft stove,  K, is
                      compared to a similarly sized conventional
                      updraft stove.  The downdraft Stove K, which
                      introduces secondary air to combustion gases
                      exiting the firebox, shows,similar;performance
                      to the updraft design.  Without secondary air,
                      emissions from Stove K would be expected to be
                      much higher.
                                      K1-4-
                                                      B2-3
           0
         20        ,          4-0
                    (Thousands)
             HEAT OUTPUT (BTU/HOUR)

                      Figure C-10
Comparison of Downdraft and Conventional Stoves
60

-------
                                      UNDERFIRE   AIR
        70
IT
6
r.9

',0
~-r
o
l/>
10
2
u
LJ
0

I
60 -
40 -
30 -
        20 -
10 -
              Stoves H and I were underfire coal/wood designs.
              Runs I 1-1 and H 1-1, H 1-3 and H 1-4 used combustion
              air brought into the firebox from directly beneath
              the fuel.  All H2 runs were made with the underfire
              air slots blocked, with air entering the firebox around
              the bottom edges of the hearth.  Although this air
              inlet was still low, emissions dropped significantly.
                         11 -1

                              H1-1
H2-6
                                     -3
             0
                             i            i            r
                            20                      4-0
                                         (Thousands)
                                HEAT OUTPUT (BTU/HOUR)
                                          Figure c-11

                                 Underfire Air Effects
                                               i
                                              60

-------
                                 2ND  AIR
o
r.0
\ — •

8
u
u
5
3'
U

i
s-u -
35 -
30 -
25 -
20 -
15 H
10 -
5 -
0 -
The effectiveness of secondary air systems is shown
below. All the stoves presented here had secondary
air systems, stoves R and C2 directed secondary air
into the main firebox from above, while stoves B, BB,
and D directed secondary air into combustion gases
at the firebox breach. A wide range of effective-
ness can be seen.

BB1&S-2 BB1-1
°2"2 02-1
J V2B1-1 R1-7
R1-6
D2~"4" D2-6
C2-3
i i i
3 20


001-6
BB1-7
B2-3

i i i
40 60
                                    (Thousands)
                             HEAT OUTPUT (BTU/HOUR)

                                    Figure C-12

                       Secondary Air Effects on Emissions

-------
     SECONDARY  AIR



'£
— *
o
to"
5
:?i>
s
6
CO
1J»,

u
i-
D
F
DL




10--


.35 -


30 -


25 -
20 -

15 -


10 -

5 -

0 -
Tests were conducted on a single
stove both with (C2) and without (Cl)
a secondary air supply which intro-
duced preheated air at the top center
area of the firebox. This stove was
especially sensitive to starting con-
ditions, so it is not clear from the
limited data whether Cl-1 and C2-3
were particularly good starts or
whether the C2 test results are as
good as they appear.

C1-4 C1~6


01-5
C2-2
C1-1

C2-3

i i i i i i
0 20 40 60
(Thousands:)
HEAT OUTPUT (BTU/HOUR)
               Figure C-13
Emissions With and Without Secondary Air

-------
SECONDARY  AIR


IT'
D
O
X
^ji;
•if
Lt:
P.
o
to
£•
LJ
LJ
§
O
i£
ii'


-4-u -1

35 -


30 -



.25 -
20 -


15 -


10 -

5 -
0 -
Ml and M2 tests all reflect the same
stove configuration, which brings
M 1—3 ' Prftheated air into the firebox through
a tube on the top of the firebox. The
stove had a 1.3 ft^ firebox and a 9 Ib
fuel load. M2 runs were conducted
with additional secondary air inlets
1 1 -I __ -I on the sides of the firebox and
several upper secondary air holes
blocked .


Ivl 1—2
M2-1
M1 -5
M2-2

M1-7

Ivl 1-6


iii ill
0 20 40 60
       [Thousands)
 HEAT OUTPUT (BTU/HOUR)
         Figure c-14
 Secondary Air Modification

-------
                                       THERM
                                                               A
o
x
"•._
10'
o

ki
LJJ
LJ


C.J
F

11'
        60 -
        50 -
        .30 -
        20 -
                                    ~ET~-2~
                                               E1-1
                                                          El and N2 stoves had primary air inlets
                                                          which were thermostatically controlled.
                                                          The thermostats on both stoves were
                                                          bimetalic springs which operated flaps
                                                          covering the  primary air inlets.  El
                                                          had a 2.5 ft3 firebox and an 18 Ib fuel
                                                          load, while N2 had a 1.7 ft3 firebox
                                                          and a 12 Ib fuel load.  The high mini-
                                                          mum burn rates with N2 makes comparison
                                                          of the stoves somewhat difficult.  El
                                                          had a low air inlet port and a low
                                                          firebox exit, which probably contributed
                                                          to the high emission rates.
                                                                                                1
                                                     N2-6
                                                           —t
                                                         \l 9 — 9

            n
                                   ._T           .            .,

                                   20                     40
                                                l'Tr,ou5'-:ind£')
                                       HEAT  OUTPUT (BTLf/HOUR)
                                                 Figure C-15
                                   Effect  of Thermostatic Controls
	T'

    '-, f!
                                                                                   -1	1

-------
                                     REFRACTORY
/—s
LL
 O
 z
 ct:
 tf
 v -••'

 !£
 .•'
 O
 "'0
 0
 ••<:
 LJ
 LJ
 C
I
35  -


30  -


25  -


20  -


15  -


10  -
5 -
         0
            D1 -2
          D2-
                    D2-4
                                  D2-6
                                             Dl runs were conducted on a steel 1.8
                                             ft3 firebox.  D2 runs had a brick and
                                             stainless steel firebox liner added,
                                             which reduced the firebox volume to 1.6
                                             ft3 and the fuel load from 12 Ibs to
                                             about 10 Ibs.  While the Insulation of
                                             the firebox is theoretically an impor-
                                             tant feature, no data is available
                                             which does not also Include changes in
                                             firebox dimensions (size) and fuel load.
           0
                        I           I           I
                       20                   4-0
                                  (Thousands)
                           HEAT OUTPUT  (BTU/HOUR)

                                    Figure  C-16
                        Effect of Refractory Materials
                                                                    I
                                                                   GO

-------
  MECH.  DRAFT


I?
~
I

to"
-^
IL
•* n
'•>
6
to
U
LJ
K
•=r
_J
r*
F
₯
CL


-t-u —

35 -


30 -



2K
G -
20 -

15 -




10 -


5 -
0 -
Stoves U and X are very similar in
design and performance. Both use
. pelletized fuel which is burned in a
cup surrounded by air supply holes
and are driven by induced draft.
Stove U feeds fuel from under the
burn cup while Stove X drops fuel in
from above. This technology appears
to be especially promising.














U 1 _ 2 U 1 - 3J 1 - 4 IJ 1 ~ 1
I i i i i • i
0 20 40 60
       (Thousands')
 HEAT OUTPUT (BTlf/HOUR)

         Figure C-17

Emissions from Pellet Stove U

-------
  MECH.  DRAFT


•JP


•:--.
to"
^
til

to
6
to
to
iti
UJ
hp
r~:
F
tr
•d
tL



•f U

35 -


30 -



25 -

20-


15 -


10 -



5 -

0 -
Stoves U and X are very similar in
design and performance. Both use
pelletized fuel which is burned in a
cup surrounded by air supply holes
and are driven by induced draft.
Stove U feeds fuel from under the
burn cup while Stove X drops fuel in
from above. This technology appears
to be especially promising.














XI -4
'x 1 _ 1 X 1 — 2^ ' ~~ •-"
ii i i i i
0 20 4-0 . 60
        (Thousands')
  HEAT OUTPUT (BTU/HOUR)

         Figure C-18

Emissions  from Pellet Stove X

-------
                                      IviEChT.    DRAFT
        4-0
'£•
n
±
Co"
Q
CO
CO

UJ
LJ
I
LL
   x
        .35 -
25 -


20 -


15 -


10


 5
                                                          VI and W2 tests were conducted as a
                                                          pellet fuel hopper mounted on a 5.1
                                                          and 2.4 ft3 stove, repectively.
                                                          Other than firebox size, the tests
                                                          were conducted under identical condi-
                                                          tions, except that VI tests had a
                                                          flue damper and W2 did not.  The
                                                          hopper unit dropped pellets into a
                                                          burn cup, which used a forced air fan.
                                                          While test results show higher
                                                          emissions than integrated pellet
                                                          units, particulate emissions are
                                                          below the 1988 Oregon catalytic
                                                          standard of 4 grams/hour.
                                                                          *'2-7
                                   I            I           I            I
                                  20                    40
                                              (Thousands)
                                      HEAT OUTPUT (BTU/HOUR)
                                                Figure C-19
                              Emissions from  a Pellet Hopper Device
                                                                                60

-------
                                  DENSIFiED   FUEL
       70
Lt
-=!'
til
c/j
(0

u
g
it
60 -
       40 -
.30  -
       20 -
1 0  -
Z 1 - 2 0
                     '.\ -'
Zl runs were conducted on a  "tradi-
tional" 2.4 ft^ firebox stove with
a fuel load of about 16 Ibs. 2.1
runs were conducted using commer-
cially available denslfied fuel logs
with the same fuel loading density.
Emissions from 2.2 runs were  signifi-
cantly (by a factor of about two)
lower.  Although only two runs were
conducted with the densifled fuel,
the emissions appear to follow the
same shape of curve as seen with the
dimensional fuel. It is not known
if the densified fuel changes the
range of heat outputs on the stove.
           0
                          20
                         40
             I ThOU Stf fids')
     MEAT  OUTPUT  (BTIf/hOJR)
               Figure C-20
    Emissions from Densified  Fuel
                                                                                 	1
                       60

-------