United States
Environmental Protection
Agency
Air and Energy Engineering     ^
Research Laboratory          ;
Research Triangle Park NC 2771 V
Research and Development
EPA/600/S2-86/100 Jan. 1987 '/'
Project  Summary
Test  Method  Evaluations  and
Emissions  Testing for  Rating
Wood  Stoves

Lawrence E. Cottone and Edward Messer
  The U.S. Environmental Protection
Agency has decided to develop a New
Source Performance Standard (NSPS)
for wood burning stoves. During the
development process several issues
must be resolved.  One  of the more
critical issues is the selection of an
emission sampling method and stove
operating  procedure.  This  report
addresses the comparison of three
candidate sampling  methods: the EPA
Modified Method 5 (MM5), the Oregon
Method 7 (OM7). and the ASTM pro-
posed Method PI 80. It also addresses
the effect emission format (g/hr, g/kg
wood burned, fjg/J heat output) has
on the intermethod correlations.
  Five stoves (i.e., two catalytic, one
noncatalytic generic, one noncatalytic
high efficiency, and one  catalytic
fireplace insert) were tested. The stoves
were nominally operated according to
the  State of Oregon's  certification
procedure. Simultaneous tests were
conducted using MM5 and OM7 in the
stove flue and MMS, OM7. and ASTM
in the (ASTM) dilution tunnel. Quality
assurance tests  using duplicate sam-
pling trains were  also conducted.
Proportional sampling, using SO2 as a
tracer gas, was conducted in the flue,
and isokinetic sampling was conducted
in the dilution tunnel.
  Results showed  good correlations
between  the total train emissions
obtained with  each method. The
strength of the correlations varied with
the  emission format; the grams per
hour format showed the strongest
correlation. POM emissions showed a
general (but weak) correlation with
total emissions  when the grams per
hour format was used; there were no
correlations when the emissions were
expressed in either of the other two
formats.
  This Project Summary was devel-
oped by EPA's Air and Energy Engi-
neering Research Laboratory. Research
Triangle  Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).

Introduction
  The purpose of this test program was
to collect  emission testing  data to
evaluate  emission measurement proce-
dures that have been applied to certifi-
cation of wood stoves and to develop an
acceptable procedure for application to
a New  Source Performance Standard
(NSPS) for the wood stove source cate-
gory.
  The primary objectives of the project
were:

• To collect emission testing  data for
  evaluating the effectiveness of the
  Oregon Department of Environmental
  Quality  (DEQ) emission sampling
  method (OM7) in distinguishing
  between low and high emissions of
  total particulate  matter, condensible
  organic matter, and polycyclic organic
  matter  (POM), using EPA Modified
  Method 5 (MMS) as the reference.

• To collect emission testing  data for
  evaluating the accuracy and precision
  of the OM7 sampling method and the
  applicability of the dilution tunnel
  emission sampling approach (as used

-------
  with the American Society for Testing
  Materials  (ASTM) home  heating
  appliance emission sampling method)
  to the OM7 sampling method.
• To  collect emission testing data for
  evaluating the representativeness of
  the ASTM emission sampling method,
  using MM5 as the reference.

  The three sampling methods (MM5,
OM7,  and ASTM) have slightly different
objectives which may account for slight
differences in  the actual  emissions
measured. MM5 was designed to collect
isokinetic  samples of particulate  and
condensible and  semivolatile organic
components. The  method captures  the
semivolatile organic fraction on a sorbent
resin.
  The OM7 method was designed for the
collection of particles and condensible
organic matter. Isokinetic sampling is not
required for sampling wood stove emis-
sions,  but  the  method does require
proportional sampling.
  The ASTM method assumes that the
particles  in the  diluted gas  stream  are
small  and behave  as a gas. The sample
rate is held constant (± 2%) throughout
the test, and gas is measured at varying
intervals  depending on the burn rate.
Since the sample is collected at basically
ambient  temperature and moisture
conditions,  the sample train filter is not
heated. The ASTM method specifies that
the sample rate be based on  the filter
loading and the filter face velocity (2 to
30 ft/min, or 0.6 to 9.1 m/min). During
these  tests, all sampling trains  were
operated  at fixed points in the stack or
dilution tunnel, as  appropriate.
  The five stoves  tested included three
catalytic stoves (a Timbereze Model 477,
a Blaze King Catalyst Stove—King Model
KEJ-1101, and a Fisher Tech IV fireplace
insert), one conventional non-catalytic
stove (Lakewood),  and one low emission
non-catalytic stove (BOSCA FS-500).

Summary of Results
  The results of the  22 test burns are
presented in Table 1. Duplicate sample
results are  included below the matching
sample run. The stove heat input was
determined from the weight of dry wood
burned per hour and the analyzed  heat
content of the fuel. The stack flow rate
was calculated using  Fc, and F factor for
wood (40 CFR 60.45, July 1985). Dilution
tunnel flows and moisture contents were
standardized to the OM7 train.
  Some of the data have been qualified
in Table 1:
• All sample runs conducted during test
  burne "Blaze King" 3 and "BOSCA"
  -3 (High Efficiency -3). These burns
  could not be sampled to completion.
  This prohibited calculations according
  to procedure OM7 and did not provide
  an emission measurement represen-
  tative of a  complete burn.  Results
  were calculated using the Fc factor.

• An isokinetic sampling conducted for
  OM7 and MM5  sample runs in the
  dilution tunnel was designated with:
  an "h" when sample  rates were over
  100% of isokinetic and a "c" when
  sampling rates were below 90%  of
  isokinetic.

• Two torn second filters in OM7 sample
  trains (located between the third and
  fourth impingers) were identified with
  an "'."

• Two test burns were not adequately
  conducted  under procedure  OM7
  stove operating guidelines. The doors
  were opened during sampling, and the
  coals and woodpieces stirred during
  two Timbereze test burns.

• Several OM7 and MM5 sample runs
  were conducted with  the heated filter
  box temperature below the  specified
  lower limit of 106°C. In all of those
  cases except BD-OM7-5, the average
  temperatures were between 93 and
  106°C.

  Total gravimetric emissions  for each
sampling method consist of the com-
bined individual analyses of the following
fractions:

• MM5
  —front half MeCI2  rinse  drydown
     residue
  —filter catch
  —XAD extraction
  —semivolatile organics (extracted
     from back-half water) by GC plus
     water drydown residue
  —semivolatile  organics (in  rinse
     solvent) by GC plus solvent drydown
     residue  condensate  extraction
     semivolatile organics by  GC plus
     drydown residue

• OM7
  —front  half  solvent  rinse  drydown
     residue
  —front filter catch
  —back half  extraction  drydown
     residue
  —back half water drydown residue
   —back half rinse drydown residue
   —back filter catch

•  ASTM
   —front  half and  between-filter
    solvent rinse drydown residue
   —front filter catch
   —backfiller catch

Table  1 also includes results of organic
sampling and  analyses conducted on
MM5  samples for total  chromatograph-
able organics (TCOs),  phenol, and POM.
TCOs  were determined on the combined
back half fractions of each MM5 sample.
POM  and phenol were analyzed  in a
combined aliquot from the front and back
half fractions of each  MM5 sample. The
POM values presented in Table 1 are the
sum of the individual values for the 21
specific compounds quantified. The
values in Table  1 do not represent total
POM, but  may indicate the  relative
change in total  POM from one burn to
another.

-------
Table 1.    Summary of Wood Stove Emission Test Results Using ORSA T Data and Fc—Grams Per Hour.
Burn No.
Timbereze
/"
2"
2&,d
Blaze King
4
4"
1
1"
5"
5d'°
2
2"
3"
Lakewood
1
3
3"
2
4
4"
Bosca
2
2"
4
1
1"
5
3"
3b,d
Fisher
2
2"
4
4"
3
1
F Factor
Heat
Input
(Btu/
hrf

17,285
23.218
23.218

1 1.607
1 1,607
20.898
20,898
31,964
31.964
53.615
53.615
6.620

17,148
22,667
22.667
29.004
139.443
139.443

10.466
10.466
21,299
28,082
28.082
64,001
1 1.447
1 1.447

10.621
10.621
18.501
18,501
30.279
38.911
Burn
Rate
(dry
kg/
hr)

0.90
1.22
1.22

0.60
0.60
1.09
1.09
1.64
1.64
2.77
2.77
0.34

0.89
1.17
1.17
1.51
17.17
17.17

0.54
0.54
1.08
1.46
1.46
3.10
0.60
0.60

0.55
0.55
0.95
0.95
1.58
2.02

f-
Stack Emissions
MM5

17.0
14.6


23.1

7.13

10.5
9.11
20.8

5.45

54.7
56.4

45.3
216.
216.

27.3

14.7
12.6

22.3
17.2
16.7'

8.54

15.9

19.7
38.9
OM7

8.01
7.55


7.99

4.59'
4.16
3.22"

10.8
10.3'
2.50

29.2'
32.1 "
35.0
26.9'
61.3


19.0

8.49
9.39
7.10
11.5
6.97


4.63

10.8

16.1
29.2
•articulate
Dilution
MM5

11.4"
13.2


10.7

10.9

13.4
15.3
24.1

15.8"

54.1
74.8
53.2
54.5
170.


18.5

28.6
32.7

68.3
56.4
59.6'

22.6C
14.9
23.9h
28.3
48.0
71.6
!

IL
u
rni
Tunnel Emissions
OM7

6.16°
4.89


7.99'
3.96
4.15

9.27'

10.7

9.43

24.4'
27.0

28.5
54.6
89.0"

8.27
12.8'
8.66'
10.7

37.2
20.2*


10.7

13.1

21.3
37.0
ASTM

4.26
3.37
0.999

4.50

2.66
2.34
4.96

8.55

2.91

27.2
33.1

22.1
97.1


12.2

9.10
9.85
8.76
20.4
19.7


4 17

6.24

13.4
30.3
Stack

6.13
4.55


12.7

2.66

4.47
3.21
6.66

2.18

19.3
26.3

18.9
109.
95.1

10.9

6.00
5.70

10.7
7.50
7.05'

2.55

4.72

5.55
12.0
Dilu-
tion
Tunnel

4.12"
3.44


4.22

3.39

3.56
3.65
6.75

2.99"

19.4
29.2
24.8
23.9
81.8


6.21

7.85
5.90

21.6
14.0
20.4'

3.25°
2.70
4.03h
3.94
6.73
12.9
Stack

0.103
0.111


0.101

0.0576

0.156
0.150
0.239

0.0223

0.411
0.444

0.782
9.11
6.46

1.74

0.351
0.199

1.02
0.402
0.252'

0.0547

0.128

0.184
0.415
tnoi
Dilu-
tion
Tunnel

0.0507C
0.0298


0.0908

0.0570

0.0996
0.112
0.258

0.0493*

0.461
0.741
0.655
1.10
3.90


0.0558

0.251
0.222

1.50
0.549
0.815'

0.08 15C
0.0641
0.0753"
0.108
0.240
0.413
ru
Stack

0.0971
0.0813


0.0319

0.0355

0.0877
0.0856
0.139

0.0316

0.148
0.228

0.337
19.3
16.9

0.0827

0.396
0.508

1 98
0.476
0.252'

0.0173

0.0415

0.0836
0.149

-------
     L Cottone is with Engineering Science, Fairfax,  VA 22030; and E. Messer is
       with Radian Corp., Research Triangle Park, NC 27709.
     Robert C. McCrillis is the EPA Project Officer (see below).
     The complete report, entitled "Test Method Evaluations and Emissions Testing
       for Rating Wood Stoves, (Order No. PB 87-119 897/AS; Cost: $18.95, subject
       to change) will be available only from:
             National Technical Information Service
             5285 Port Royal Road
             Springfield, VA  22161
             Telephone:!. 703-487-4650
     The EPA Project Officer can be contacted at:
             Air and Energy Engineering Research Laboratory
             U.S. Environmental Protection Agency
             Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
                                                                                               i i e» oPPT''M  M*
                                                                                               U.O.\-/i r«O»r-vl_ I»lf»
Official Business
Penalty for Private Use $300
EPA/600/S2-86/100
  0000329   PS
  U  S  ENVIR  PROTECTION  AGENCY
  CHICAGO

-------