Stationary
Reciprocating
Internal
Combustion
Engines
Technical
Support
Document
for
NOx
SIP
Call
October
2003
Doug
Grano/
Bill
Neuffer
EPA,
OAR,
OAQPS,
AQSSD,
OPSG
1
1Large,
as
defined
in
the
NOx
SIP
call
(
63
FR
57356,
October
27,
1998),
means
an
IC
engine
which
emitted,
on
average,
greater
than
1.0
ton
of
NOx/
day
during
the
1995
ozone
season.

2Alternative
Control
Techniques
(
ACT)
document,
"
NO
x
Emissions
from
Stationary
Reciprocating
Internal
Combustion
Engines,"
(
ACT
document
for
IC
engines)
EPA­
453/
R­
93­
032,
July
1993,
page
3­
15.
The
ACT
documents
were
required
by
section
183(
c)
of
the
Clean
Air
Act
Amendments
of
1990
and
subject
to
public
review
prior
to
publication.

3"
Retrofit
NOx
Control
Technologies
for
Natural
Gas
Prime
Movers,"
Gas
Research
Institute,
March
1994,
GRI­
94/
0329,
page
2­
4,
(
1994
GRI
report).

4See,
for
example,
data
from
EPA's
AP­
42,
Emission
Factors
document,
Table
3.2­
1,
10/
96.

5See
NOx
SIP
call
final
rule
and
support
material
(
63
FR
57356,
October
27,
1998).

6"
Highly
cost­
effective
controls"
are
defined
in
the
NOx
SIP
call
as
controls
which
are
less
than
$
2000/
ton
of
ozone
season
NOx
reduction
in
1990
dollars
(
63
FR
57356,
October
27,
1998).

7The
discussion
below
uses
"
grams/
brake
horsepower­
hour"
or
g/
bhp­
hr
rather
than
lbs/
mmBtu
since
the
former
is
the
convention
for
the
industry.
The
uncontrolled
estimate
of
3.0
lbs/
mmBtu
(
from
AP­
42,
October
1996)
corresponds
to
about
11.3
g/
bhp­
hr.
The
1993
ACT
document
for
IC
engines
estimates
average
uncontrolled
emissions
at
5.13
lb/
mmBtu
or
16.8
g/
bhp­
hr.
Introduction
Large1
stationary
reciprocating
internal
combustion
engines
(
IC
engines)
are
primarily
used
in
pipeline
transmission
service
and
some
are
used
in
field
storage
pumping
operations.
Gas
turbines
are
also
used
in
these
operations.
On
a
capacity
basis
the
IC
engines
and
turbines
in
pipeline
transmission
service
are
about
evenly
divided.
2,3
The
uncontrolled
emission
rate
from
IC
engines
is
about
ten
times
greater
than
the
uncontrolled
emission
rate
for
gas
turbines.
4
That
is,
uncontrolled
NOx
emissions
from
large
IC
engines
are
greater
than
3.0
lbs/
mmBtu
while
uncontrolled
NOx
emissions
from
gas
turbines
are
about
0.3
lbs/
mmBtu.

In
the
NOx
SIP
call,
EPA
determined
that
NOx
emissions
from
large
gas
turbines
(
and
large
boilers)
can
be
decreased
by
highly
cost­
effective
controls
to
an
average
emission
rate
of
0.15­
0.17
lbs/
mmBtu5.
As
part
of
the
NOx
SIP
call
rulemaking,
EPA
stated
that
highly
costeffective
controls6
are
available
to
reduce
emissions
from
large
IC
engines
by
90%
from
uncontrolled
levels
(
i.
e.,
to
about
0.3
lbs/
mmBtu)
7.
The
DC
Circuit
Court
in
a
March
3,
2000
2
8Federal
Register
of
March
2,
2000
(
65
FR
11222).

9"
NOx
Emissions
Control
Costs
for
Stationary
Reciprocating
Internal
Combustion
Engines
in
the
NOx
SIP
Call
States"
prepared
by
Pechan­
Avanti
Group
for
EPA,
August
11,
2000
(
Pechan
IC
engines
report).

10
Annual
(
capital
and
operating)
costs
in
1990
$
per
ozone
season
tons
reduced.
For
SCR
and
NSCR,
the
annual
operating
costs
are
for
the
ozone
season
only.
LEC
controls
are
assumed
to
operate
year­
round,
thus,
year­
round
operating
costs
are
included.
For
comparison
to
other
recent
EPA
rulemakings,
the
costs
can
be
escalated
to
1997
$
using
a
factor
of
1.21,
resulting
in
$
629­
664/
ton.

11"
Stationary
Reciprocating
Internal
Combustion
Engines:
Updated
Information
on
NOx
Emissions
and
Control
Techniques,"
EC/
R
Incorporated,
September
1,
2000.
decision
ruled
that
EPA
had
not
provided
adequate
notice
and
opportunity
to
comment
on
the
IC
engines
control
level
EPA
used
to
determine
the
State
NOx
budgets
for
the
final
rule.
In
the
February
22,
2002
proposed
rulemaking,
EPA
proposed
that
highly
cost­
effective
controls
are
available
to
reduce
emissions
from
large
IC
engines
by
82­
91%
(
see
9­
5­
00
TSD).

In
the
October
27,
1998
final
NOx
SIP
call
rule,
EPA
identified
about
300
large
IC
engines.
Subsequently,
EPA
received
information
from
commenters
seeking
to
make
changes
to
the
emissions
inventory.
The
EPA
made
corrections
and
now
includes
180
IC
engines
in
its
final
NOx
SIP
call
budget8.
The
vast
majority
are
natural
gas­
fired
engines.

An
August
2000
report
by
the
Pechan­
Avanti
Group
estimates
the
control
costs
and
NO
x
emission
reductions
for
large
IC
engines
affected
under
the
NOx
SIP
Call.
The
report
provides
information
about
the
universe
of
potentially
affected
IC
engines,
control
cost
modeling
methods,
scenario
analyses,
and
caveats
and
uncertainties
associated
with
this
analysis.
9
For
the
control
range
of
82­
93%,
the
report
estimates
the
average
cost
effectiveness
to
be
$
520­
549
per
ton.
10
A
September
2000
report
by
EC/
R
also
contains
estimates
of
the
control
costs
and
NO
x
emission
reductions
for
large
IC
engines.
The
EC/
R
report
estimates
the
average
cost
effectiveness
for
IC
engines
2,000­
8,000
hp
to
be
$
420­
840
per
ton.
11
Large
IC
Engines
Except
Natural
Gas­
Fired
Lean­
Burn
3
12"
Stationary
Reciprocating
Internal
Combustion
Engines:
Updated
Information
on
NOx
Emissions
and
Control
Techniques,"
EC/
R
Incorporated,
September
1,
2000
(
EC/
R
report
on
IC
engines).

13ACT
document
for
IC
engines,
Tables
2­
2
and
2­
12.
In
the
initial
1998
NOx
SIP
call
budget
calculation,
EPA
divided
IC
engines
into
4
categories
and
assigned
a
90
percent
emissions
decrease,
on
average,
to
each
category.
This
reflected
non­
selective
catalytic
reduction
(
NSCR)
for
rich­
burn
engines
and
selective
catalytic
reduction
(
SCR)
for
diesel
and
dual­
fuel
engines.
For
all
large
IC
engines,
except
natural
gasfired
variable
load
lean­
burn
engines
(
see
discussion
below),
EPA
continues
to
believe
that
90%
control
is
achievable
through
NSCR
or
SCR
and
is
highly
cost­
effective.
This
is
demonstrated,
for
example,
in
the
1993
ACT
document
for
IC
engines
and
in
the
9­
1­
00
EC/
R
report
which
updates
information
on
NOx
emissions
and
control
techniques
for
IC
engines.
12
In
addition,
the
following
sources
provide
supporting
information
(
see
docket
A­
96­
56):

*
"
NOx
Reduction
Technology
for
Natural
Gas
Industry
Prime
Movers,"
Acurex
Corporation
for
Gas
Research
Institute,
August
1990.

*
"
Retrofit
NOx
Control
Technologies
for
Natural
Gas
Prime
Movers,"
section
4,
Gas
Research
Institute,
March
1994,
GRI­
94/
0329.

*
"
Assessment
of
Control
Technologies
for
Reducing
Nitrogen
Oxide
Emissions
from
Non­
Utility
Point
Sources
and
Major
Area
Sources,"
Final
Ozone
Transport
Assessment
Group
(
OTAG)
Policy
Paper,
July
1996;
Chapter
5,
Appendix
C,
to
the
OTAG
Final
Report,
http://
www.
epa.
gov/
ttn/
rto/
otag/
index.
html.

*
"
Emission
Control
Technology
for
Stationary
Internal
Combustion
Engines,"
Status
Report,
Manufacturers
of
Emission
Controls
Association,
July
1997.

*
"
California
Environmental
Protection
Agency/
Air
Resources
Board
­
Determination
of
Reasonably
Available
Control
Technology
and
Best
Available
Retrofit
Control
Technology
for
Stationary
Spark­
Ignited
Internal
Combustion
Engines,"
November
2001
*
CAPCOA/
ARB
­
"
Determination
of
Reasonably
Available
Control
Technology
and
Best
Available
Retrofit
Control
Technology
for
Stationary
Internal
Combustion
Engines
­
Draft,"
December
3,
1997
Natural
Gas­
Fired
Rich­
Burn
Engines
Non­
selective
catalytic
reduction
(
NSCR)
provides
the
largest
NO
x
percent
reduction
of
all
the
highly
cost
effective
technologies
considered
in
the
ACT
document
as
it
is
capable
of
providing
a
90
to
98
percent
reduction
in
NO
x
emissions.
13
The
EC/
R
report
on
IC
engines
4
14EC/
R
report
on
IC
engines,
section
4.3.4.

15Telephone
records
by
Bill
Neuffer,
EPA,
dated
5­
19­
00
and
5­
24­
00;
conversations
with
a
regulatory
agency
representative,
an
operator
of
the
control
equipment
and
an
equipment
vendor.

16ACT
document
for
IC
engines,
Tables
2­
8,
2­
14
and
2­
15.

17"
Emission
Control
Technology
for
Stationary
Internal
Combustion
Engines,"
Status
Report
by
Manufacturers
of
Emission
Controls
Association,
July
1997,
page
7
(
1997
MECA
report).

18"
CAPCOA/
ARB
­
Draft
­
Determination
of
Reasonably
Available
Control
Technology
and
Best
Available
Retrofit
Control
Technology
for
Stationary
Internal
Combustion
Engines,"
December
3,
1997,
page
29.

19EC/
R
report
on
IC
engines,
section
4.2.4.
states
that
95
percent
control
is
generally
achievable
through
the
use
of
NSCR
on
rich­
burn
IC
engines.
14
The
time
required
from
cost
proposal
to
field
installation
of
NSCR
is
less
than
11
months.
15
Diesel
and
Dual
Fuel
Engines
For
diesel
and
dual
fuel
engines,
SCR
provides
the
largest
NO
x
reduction
of
all
highly
cost
effective
technologies
considered
in
the
1993
ACT
document.
It
is
reported
to
provide
an
80­
90
percent
reduction
in
NO
x
emissions.
16
More
recent
reports
state
that
NO
x
emissions
can
be
reduced
by
90%
or
more
by
SCR.
17,
18,
19
Therefore,
EPA
estimates
NOx
reductions
for
these
engines
at
90%
on
average.
The
EPA
estimates
the
diesel/
dual
fuel
IC
engines
are
a
very
small
part
of
the
large
IC
engines
population
in
the
NO
x
SIP
call.
There
are
only
5
large
diesel
IC
engines
identified
in
the
SIP
call
jurisdictions,
some
of
which
may
be
capable
of
dual
fuel
operation.

Natural
Gas­
Fired
Lean­
Burn
IC
Engines
Uncontrolled
Emission
Rate
The
EPA
examined
data
on
large
natural
gas
fired
lean
burn
engines
obtained
from
the
5
pipeline
industry,
collected
by
the
Agency,
and
contained
in
the
ACT
document.
These
include
data
from
large
natural
gas
fired
lean
burn
engines
covered
by
the
SIP
call.
The
EPA
believes
the
data
supports
the16.8
g/
bhp­
hr
value
proposed
on
February
22,
2002,
as
described
below.

One
of
the
data
sets
that
supports
the
16.8
g/
hp­
hr
level
is
additional
data
developed
by
pipeline
industry
members
that
is
based
on
a
survey
of
LEC
retrofit
installation
in
SIP
call
States.
In
a
November
20,
2000
letter
from
Tennessee
Gas
Pipeline
&
Transcontinental
Gas
Pipe
Line
to
the
Ozone
Transport
Commission,
survey
data
presented
in
Attachment
A
of
the
letter
include
both
pre­
LEC
and
post­
LEC
data
for
86
engines
in
NOx
SIP
call
States.
Most
of
the
engines
are
2000
hp
or
greater.
Table
1
of
the
letter
summarizes
the
data
and
states
that
the
average
uncontrolled
NOx
emissions
level
for
these
86
engines
is
16.8
g/
bhp­
hr.
The
range
of
uncontrolled
values
is
7.0­
25.8
g/
bhp­
hr.
Considering
only
those
engines
greater
than
or
equal
to
2,000
hp,
there
are
66
engines
with
an
average
uncontrolled
emissions
rate
of
18.2
g/
hp­
hr
(
see
table
below).

From
Attachment
A
(
engines
>
or
=
2,000
hp):
Location
Engine
Uncontrolled(
g/
bhp­
hr)
AL­
Station
110
C­
B
V­
250­
16
(
2)
23.9
MD
­
Station
190
Clark
TCV­
10;
16
14.2,
12.2
NJ
­
Station
505
I­
R
412
KVS
(
8)
21.8
NY­
Station
237
Clark
TCV­
10
9.0
NY
­
Station
241
Clark
TLA­
10(
2
engines
at)
7.0
NY­
Station
224
I­
R
KVS
412(
4)
16.0
NY
­
Station
237
I­
R
KVS­
412
(
2)
16.0
PA­
Station
219
C­
B
GMV­
10(
2)
16.0
PA
­
Station
307
Clark
TCV­
10
9.0
PA­
Station
307
I­
R
KVS­
412
(
4)
16.0
PA
­
Station
219
C­
B
V­
250­
16
11.0
PA­
Station
200
Clark
TLA­
6
(
4)
14.5
PA
­
Station
200
Clark
TCV­
10(
2)
9.0
PA­
Station
200
Clark
TCV­
16
12.0
PA­
Station
515
C­
B
GMWC­
10(
3)
25.8
PA
­
Station
535
I­
R
36
KVS
18.6
PA
­
Station
520
I­
R
412
KVS
(
5)
22.4
PA­
Station
535
I­
R
512
KVS
(
3)
17.8­
2;
17.2
PA­
Station
515
C­
B
V­
250­
10
(
2)
23.3
PA­
Station
195
C­
B
V­
250­
12
(
2)
18.1
TN
­
Station
87
C­
B
V­
250­
16
11.0
TN
­
Station
2101
I­
R
KVS­
412
16.0
TN­
Station
2101
C­
B
V­
250­
8
18.0
VA
­
Station
180
Clark
TCV­
10
(
3)
12.0
VA
­
Station
185
I­
R
412
KVS
(
10)
22.4
6
Attachment
B
to
the
same
11­
20­
00
letter
summarizes
pre­
LEC
and
post­
LEC
data
for
20
engines.(
see
table
below).
Fourteen
of
the
20
engines
are
2,000
hp
or
greater.
The
letter
states
that
the
average
uncontrolled
NOx
emissions
for
the
20
engines
is
14.1
g/
bhp­
hr
and
the
range
of
uncontrolled
values
is
7.0­
18.0
g/
bhp­
hr.
Considering
only
the
engines
from
this
data
set
greater
than
or
equal
to
2,000
hp,
the
average
uncontrolled
emissions
for
these
engines
is
also14.1
g/
hphr

From
Attachment
B
(
engines
>
or
=
2,000
hp):
Station
Engine
Uncontrolled(
g/
hp­
hr)
NY­
Station
237
Clark
TCV­
12
9.0
NY­
Station
241
Clark
TLA­
10
(
2
engines)
7.0
NY­
Station
237
I­
R
KVS­
412
16.0
NY­
Station
224
"
16.0
NY­
Station
237
"
16.0
NY­
Station
224
"
16.0
PA­
Station
307
"
(
4
engines)
16.0
PA­
Station
219
C­
B
V­
250­
16
11.0
TN­
Station
87
C­
B
V­
250­
16
18.0
TN­
Station
2101
C­
B
V­
250­
8
18.0
Consolidated
Natural
Gas
Service
Company,
a
major
pipeline
company,
also
sent
a
letter,
dated
11/
22/
00
to
the
Ozone
Transport
Commission
(
OTC)
concerning
the
OTC's
development
of
a
set
of
model
NOx
rules.
The
attachment
to
Dominion's
11­
22­
00
letter
to
OTC,
contains
uncontrolled
and
RACT
emission
rates
for
62
engines
retrofit
with
LEC
(
see
Table
1).
The
average
uncontrolled
emission
rate
taken
into
consideration
all
62
engines
from
this
data
set
is
17.6
g/
bhp­
hr.
Considering
the
average
emissions
for
each
of
the
18
models
gives
17.2
g/
bhp­
hr.
Although
these
engines
are
"
major"
sources
since
they
are
subject
to
RACT,
it
is
not
clear
if
all
are
"
large"
engines
with
respect
to
the
NOx
SIP
call.

Table
1.
Uncontrolled
Emissions
­
Dominion's
11­
22­
00
Letter
Number
of
Engines
Engine
Model
Uncontrolled
NOx
emissions
(
g/
hp­
hr)

2
Ajax
DPC­
600
15.5
5
Clark
HBA­
5T
23
7
6
Clark
HLA­
8
27
5
Clark
TLA­
6
16
3
Clark
TLA­
6
16
2
Clark
TLA­
6
16
5
Clark
TLA­
6
16
2
Clark
TCV­
10
16
3
Clark
TLA­
10
16
4
Clark
TCV­
10
16
2
Cooper
14W330
13
5
Cooper
GMVC­
6
11
3
IR
36
KVS­
FT
20
1
IR
48
KVS­
ET
20
3
IR
103
KVG­
ML
16
3
IR
104
KVG­
LL
16
3
IR
512
KVS­
FT
16
5
IR
512
KVS­
ET
20
Total:
62
engines
Average:
17.6
EPA
collected
additional
test
data
to
better
determine
controlled
and
uncontrolled
emission
levels
from
the
current
population
of
large
engines
in
the
NOx
SIP
call
area.
The
data
were
placed
in
the
docket
and
the
uncontrolled
emission
rate
data
are
summarized
in
Table
2.
The
average
uncontrolled
NOx
level
from
this
set
of
42
test
values
is
16.7
g/
bhp­
hr,
nearly
identical
to
the
proposed
level
of
16.8
g/
bhp­
hr.

Table
2.
Uncontrolled
Emissions
­
Additional
Test
Data
­
SIP
Call
Area
Engine
Model
Uncontrolled
NOx
emissions
(
g/
hp­
hr)
Location
Reference
CB
GMW
20.6
GA
Transco
Station
120
5­
22­
02
fax
from
EPA
Region
4
8
CB
GMW
20.1
(
avg.
6
tests)
TX
Transco
Station
40
6­
3­
02
e­
mail
from
TNRCC
CB
GMW­
6TF
17.4
KY
Texas
Gas
4­
10­
02
e­
mail
from
Jon
Trout
CB
GMW­
8
14.5
TN
Tenneco
Station
87
6­
2­
02
e­
mail
from
EPA
Region
4
CB
V­
250
18.3
PA
Transco
Station
195
6­
28­
02
e­
mail
from
State
of
PA
CB
V­
250
23.3
PA
Transco
Station
515
6­
28­
02
e­
mail
from
State
of
PA
CB
8V­
250
16.9
TN
MW
Station
2101
6­
2­
02
e­
mail
from
EPA
Region
4
CB
16V­
250
18.3
TN
Tenneco
Station
87
6­
2­
02
e­
mail
from
EPA
Region
4
CB
16V­
250
23.9
AL
Tenneco
Station
110
5­
22­
02
e­
mail
from
EPA
Region
4
CB
GMWA
13.6
KY
Tenn.
Gas
Jefferson
Co.
4/
10/
02
email
from
Jon
Trout
CB
GMWA­
8
16.0
TX
Vidor
6­
3­
02
e­
mail
from
TNRCC
CB
GMWA­
8
20.9
TN
Coastal
Cottage
Grove
1­
5­
01
letter
Coastal
to
State
of
TN
CB
GMWC
25.8
PA
Transco
Station
515
6­
28­
02
e­
mail
from
State
of
PA
CB
GMWC­
10
32.4
TN
Tenneco
Station
87
6­
2­
02
e­
mail
from
EPA
Region
4
and
2­
21­
95
letter
from
Tenneco
to
TN
CB
GMVA
18.2
CA
Mobil
Rincon
EC/
R
9­
00
report,
p.
30
CB
W330
12.5
NY
Tenn.
Gas
Station
241
5­
29­
02
e­
mail
from
EPA
Region
2
Clark
HLA
27
PA
Dominion
South
Bend
6­
28­
02
e­
mail
from
State
of
PA
Clark
HBA­
8T
8.4
(
avg
of
7
tests)
MD
Transco
Station
190
1995
test
data
sent
by
Maryland
­
9/
02
9
Clark
TCV­
10
TCV­
16
8.4
11.3
Transco
Station
200
6­
28­
02
e­
mail
from
State
of
PA
Clark
TCV­
12
13
NY
Station
237
5­
29­
02
e­
mail
from
EPA
Region
2
(
OEM
estimate)

Clark
TCVC­
20
10.1
TN
ANR
Cottage
Grove
6­
2­
02
e­
mail
from
EPA
Region
4
Clark
TCVD­
16
12.8
13.0
TN
Coastal
Cottage
Grove
6­
1­
02
e­
mail
from
EPA
Region
4
and
10­
5­
00
letter
Coastal
to
TN
Clark
TLA
9.6
10­
92
Acurex
report
to
GRI
Clark
TLA
13
NY
Tenneco
Syracuse
5­
29­
02
fax
from
EPA
Region
2
Clark
TLA
9.8
MI
Consumers
Energy
Oversiel
6­
7­
02
e­
mail
from
State
of
Michigan
Clark
TLA
13.4
13.1
16.1
15.7
NY
Algonquin
Stony
Point
(
4
engines)
5­
24­
02
fax
from
EPA
Region
2
Clark
TLA
13.3
11.5
15.0
MD
Transco
Station
190
(
3
engines)
Information
sent
by
Maryland
­
8/
02
IR
KVS­
412
8.1
10­
92
Acurex
report
for
GRI
IR
KVS
24.4
PA
Transco
Station
520
6­
5­
91
letter
from
Transco
IR
KVS
25
NY
Tenneco
Clymer
Station
5­
29­
02
fax
from
EPA
Region
2
IR
KVS
25
NY
Tenneco
Clifton
Springs
5­
29­
02
fax
from
EPA
Region
2
IR
KVS
24.8
(
1
test
result
for
2
engines)
TN
Tenneco
Station
2101
6­
02­
02
e­
mail
from
EPA
Region
4
and
2­
21­
95
letter
from
Tenneco
to
TN
IR
KVS
19.4
TX
Vidor
Station
6­
3­
02
e­
mail
from
TNRCC
10
20"
Stationary
Reciprocating
Internal
Combustion
Engines:
Updated
Information
on
NOx
Emissions
and
Control
Techniques,"
EC/
R
Incorporated,
September
1,
2000
(
EC/
R
report
on
IC
engines).
IR
KVR
8.2
TX
Motiva
6­
3­
02
e­
mail
from
TNRCC
IR
KVT­
512
21.4
TN
Tenneco
Station
2101
6­
02­
02
e­
mail
from
EPA
Region
4
and
2­
21­
95
letter
from
Tenneco
to
TN
16.7
Average
Uncontrolled
emissions
data
are
also
reported
in
chapter
3
of
the
EC/
R
report,
20
as
summarized
below.
The
data
show
a
wide
range
of
values,
due
in
part
to
the
inclusion
of
some
engines
considered
by
the
EC/
R
report
as
being
controlled.

A
1994
Gas
Research
Institute
(
GRI)
report
indicated
separate
emission
levels
for
2
stroke
(
12.5
g/
bhp­
hr)
and
4
stroke­
engines
(
13.2
g/
bhp­
hr).
Test
results
for
2
stroke
engines
range
from
2­
29
g/
bhp­
hr.
For
4­
stroke
engines,
results
range
from
1­
25
g/
bhp­
hr.
The
report
noted
that
the
higher
end
25­
29
g/
bhp­
hr
was
representative
of
the
older
uncontrolled
engines
(
these
are
the
engines
most
likely
affected
by
the
SIP
Call).
Engines
equipped
with
turbochargers
and
intercoolers
as
original
design
features
typically
emit
7­
15
g/
bhp­
hr.
The
lower
end
of
the
range
often
reflects
the
newer
lean
burn
engines
which
achieve
1­
2
g/
bhp­
hr.
Thus,
the
average
emission
levels
presented
in
this
GRI
report
were
calculated
including
some
engines
considered
controlled
for
purposes
of
the
EC/
R
report.

In
the
AP­
42
(
10/
96)
document,
uncontrolled
emissions
are
reported
for
2­
stroke
engines
at
10.9
g/
bhp­
hr
and
for
4­
stroke
at
11.8
g/
bhp­
hr.
This
report
uses
many
of
the
same
test
data
references
as
1994
GRI
report.
The
EC/
R
report
states
that
it
appears
likely
the
uncontrolled
data
include
test
reports
from
newer
lean­
burn
engines
that
would
be
considered
controlled.

In
the
AP­
42
(
1997
draft
revision)
document,
uncontrolled
emissions
for
2
stroke
are
12.2
g/
bhp­
hr
and
15.0
g/
bhp­
hr
for
4­
stroke.
This
is
based
on
38
tests
for
2­
stroke
and
18
tests
for
4­
stroke.
The
EC/
R
report
notes
that
some
lean
burn
engines
in
this
database
are
actually
controlled
emissions
by
LEC
technology.

A
1996
GRI
report
includes
data
on
six
2
stroke
engines
representing
5
models.
Each
engine
was
tested
2­
5
times.
The
2­
stroke
engine
averages
ranged
from
4.9
­
20.8
g/
bhphr
and
the
4­
stroke
engine
averages
ranged
from
7.0
g/
bhp­
hr
­
22.0
g/
bhp­
hr.
The
test
data
were
more
concentrated
towards
the
lower
end
of
each
range.

A
1998
GRI
report
includes
data
from
a
Cooper
Z­
330
engine
that
had
not
been
11
21"
NOx
Control
for
Two­
Cycle
Pipeline
Reciprocating
Engines"
prepared
by
Arthur
D.
Little,
Inc.
for
GRI,
December
1998,
figure
1­
1.
retrofitted
with
Clean
Burn
to
be
up
to
24
g/
bhp­
hr.
21
Emissions
from
2
other
models
were
reported
to
range
from
6­
13
g/
bhp­
hr
and
11.5
for
another
model.

Uncontrolled
1995
test
data
from
a
PG&
E
site
for
2
Cooper
Bessemer
W­
330
models
is
reported
to
be
18.9
and
16.7
g/
bhp­
hr.
(
EC/
R
reference
9,
page3­
14,
letter
and
attachments
from
Carol
Burke,
PG&
E
to
W.
Neuffer
­
2/
3/
00.)

Test
data
from
So
Cal
Gas
is
reported
for
2
Ingersoll
Rand
412KVS
models
to
be
21.4
and
17.0
g/
bhp­
hr.
(
Reference
page
3­
4,
EC/
R
report.)

A
1990
GRI
report
stated
uncontrolled
emissions
for
lean
and
rich
burn
to
range
from
7­
26
g/
bhp­
hr.

A
1992
paper
prepared
by
Cooper
for
Society
of
Petroleum
Engineers
states
that,
prior
to
regulation,
for
both
lean
and
rich
burn
engines,
NOx
emissions
range
from
10­
20
g/
bhp­
hr.

A
1997
Manufacturer
of
Emission
Control
Association
report
states
that
typical
NOx
emissions
for
engines
that
operate
slightly
lean
of
stoichiometric
is
18
g/
bhp­
hr.

A
1994
Oil
and
Gas
Journal
article
on
natural
gas
compressor
station
engines
indicates
that
typically
emissions
are
15
g/
bhp­
hr,
for
both
lean
and
rich
burn
engines.

During
a
visit
to
a
So
Cal
Gas
plant,
a
representative
of
the
plant
stated
that
for
a
DeLaval
HVA16C
engine,
uncontrolled
emissions
were
28
g/
bhp­
hr
prior
to
installing
LEC.

Product
literature
from
Ajax
Superior
Division
of
Cooper
Energy
indicates
uncontrolled
emissions
from
an
Ajax
2­
stroke
lean­
burn
engine
(
110
­
720
bhp)
range
from
3.0­
9.5
g/
bhp­
hr
and
from
a
Superior
4­
stroke
lean­
burn
engine
(
825­
2650
bhp)
range
from
15.0
­
22.1
g/
bhp­
hr.

As
described
in
the
ACT
document,
uncontrolled
emission
levels
were
provided
to
EPA
by
several
engine
manufacturers.
These
emission
levels
were
tabulated
and
averaged
for
engines
with
similar
power
ratings.
Most
manufacturers
provided
emission
data
only
for
current
production
engines,
but
some
included
older
engine
lines
as
well.
For
lean
burn
engines,
the
average
ranges
from
7.9­
18.6
g/
hp­
hr.
The
7.9
g/
hp­
hr
represents
the
smallest
engine
category
and
is
considerably
lower
than
all
the
other
lean
burn
engines
size
categories.
As
can
be
seen
from
the
data
below,
there
is
considerable
agreement
in
the
value
for
the
larger
engines,
with
a
average
range
of
16.5­
18.6.
This
is
significant
because
the
SIP
call
specifically
addresses
large
engines.
12
22From
Table
4­
1
­
ACT
Document.

23For
large
lean­
burn
IC
engines
in
the
NOx
SIP
Call
states,
2­
stroke
engines
represent
83%
of
the
total
large
engines
and
85%
of
the
total
large
engine
horsepower.
(
From
INGAA's
From
Table
4­
1
­
ACT
Document
Lean
burn
engines
(
g/
bhp­
hr)
22
Size
No.
of
engines
Highest
Lowest
Average
(
HP)
in
data
base
0­
400
7
17.5
3.0
7.9
401­
1,000
17
27.0
 
15.5
18.6
1,001­
2,000
43
27.0
14.0
17.8
2,001­
4,000
30
27.0
10.0
17.2
4,001+
25
17.5
10.0
16.5
There
are
several
reasons
to
use
the
ACT
document
data:

*
Using
the
applicable
ACT
document
rather
than
AP­
42
is
consistent
with
our
treatment
of
other
non­
EGU
source
categories,
including
glass,
process
heaters,
iron
&
steel,
and
other
industrial
source
categories
in
the
NOx
SIP
call
rulemaking.

*
The
ACT
document
provides
a
comprehensive
look
at
the
IC
engine
class
and
has
the
advantage
of
using
a
consistent
data
set
for
uncontrolled
emissions,
costs,
and
controls.

*
If
we
used
AP­
42
uncontrolled
numbers,
it
would
be
logical
to
use
the
AP­
42
controlled
numbers.
However,
the
AP­
42
controlled
data
set
is
limited
in
terms
of
technologies
considered,
costs,
and
expected
decreases
in
emissions.

*
The
ACT
document
uses
a
large
data
set
from
which
to
draw
conclusions.

*
ACT
test
data
are
available
in
several
horsepower
size
categories;
this
is
important
since
EPA
chose
to
not
calculate
emission
reductions
from
the
smaller
IC
engines.
The
16.8
g/
bhp­
hr
appears
to
be
more
representative
of
larger
engines,
which
are
the
engines
affected
by
the
NOx
SIP
call.

EPA
also
examined
the
available
data
separately
for
2­
and
4­
stroke
engines.
As
shown
in
Table
3,
the
test
data
for
the
large
IC
engines
in
the
SIP
call
area
indicate
uncontrolled
levels
of
16.4
and
18.9,
respectively,
for
the
2­
and
4­
stroke
engines.
Using
information
from
the
pipeline
industry
that
about
85%
of
the
engines
in
the
SIP
Call
area
are
2­
stroke,
the
weighted
average
of
the
16.4
and
18.9
values
is
16.8,
identical
to
EPA's
proposed
value.
23
EPA
believes
these
data
13
April
22,
2002
comments,
pages
2
and
10.)
support
the
16.8
value
proposed
by
EPA.

Table
3.
Uncontrolled
Emissions
­
2­
Stroke;
4­
Stroke
Data
Source
2­
Stroke
Average
Emission
Rate
(#
engine
tested)
4­
Stroke
Average
Emission
Rate
(#
engine
tested)

Attachment
A
15.7
g/
hp­
hr
(
28)
19.7
g/
hp­
hr
(
37)

Attachment
B
11.7
"
(
6)
16.0
"
(
8)

Dominion
17.6
"
(
44)
18.0
"
(
18)

Additional
Tests
16.1
"
(
35)
20.1
"
(
9)

Totals
16.4
"
(
114)
18.9
"
(
76)

In
addition,
EPA
reviewed
the
data
used
to
update
AP­
42.
In
order
to
focus
on
the
type
of
engines
addressed
in
the
NOx
SIP
call,
EPA
examined
test
data
from
those
engines
greater
than
2,000
HP
operating
at
greater
than
90%
load.
As
a
result,
the
average
emission
rate
is:
12.2
grams.
Further,
if
we
remove
2
extremely
low
values
 
which
probably
represent
reduced
engine
emissions
due
to
turbocharging
[
2.2
and
6.3
grams]
 
the
average
is
14.9
grams.
The
group
of
large
engines
in
this
database
represents
only
2
engine
models
and
8
tests;
both
models
are
4­
stroke
engines.
The
data
are
summarized
below
(
NO
x
emissions
in
this
database
were
given
in
ppm
NO
2
@
x%
oxygen;
values
were
converted
to
ppm
NO
2
@
15%
oxygen
and
then
converted
to
g/
hp­
hr
by
dividing
by
70).

The
engines
considered
were:
29.33x
­
Cooper
Bessemer
LSV­
16
­
4,200
HP
­
13.1
g/
hp­
hr
29.34x
­
"
"
"
"
­
12.2
g/
hp­
hr
29.35x
­
"
"
"
"
­
6.3
g/
hp­
hr
29.36x
­
"
"
"
"
­
2.2
g/
hp­
hr
29.37x
­
"
"
"
"
­
9.6
g/
hp­
hr
29.38x
­
"
"
"
"
­
11.2
g/
hp­
hr
29.40x
­
Ingersoll­
Rand
KVS
­
412
­
2,000
HP
­
20.8
g/
hp­
hr
29.41x
­
"
"
"
"
­
2,000
HP
­
22.3
g/
hp­
hr
The
data
in
the
7­
00
AP­
42
update
do
not
differentiate
between
uncontrolled
lean­
burn
engines
and
engines
that
may
be
turbocharged.
Thus,
the
average
"
uncontrolled"
emissions
reported
may
include
some
engines
with
lower
NOx
emissions
due
to
the
turbocharging.
See
footnotes
"(
a)"
to
Tables
3.2­
1
and
3.2­
2
in
the
7­
00
AP­
42
document
identifying
this
concern.
It
14
24For
example,
November
30,
1998
letter
from
Lisa
Beal,
INGAA,
to
docket
A­
98­
12
(
docket
#
III­
D­
53)
and
February
16,
1999
memo
from
Lisa
Beal,
INGAA,
to
Tom
Helms,
EPA.

25EC/
R
report
on
IC
engines,
section
4.2.

26For
example,
December
1,
1998
letter
from
INGAA
to
EPA
docket,
February
16,
1999
memo
from
INGAA
to
Tom
Helms,
EPA,
and
April
26,
2002
comment
letter
from
Kinder
is
important
to
note
that
essentially
all
modern
engines
above
300kW
are
turbocharged
to
achieve
higher
power
densities
(
Energy
Nexus
Group,
Inc,
p16,
Feb.
2002).
The
effect
of
turbocharging
is
to
increase
the
air/
fuel
ratio,
which
will
lower
the
NOx
emissions.
Thus,
the
AP­
42
data
(
2002
document)
appear
to
reflect
a
newer
engine
population
with
a
lower
average
emission
rate
which
may
not
be
representative
of
the
older
SIP
call
population.

In
summary,
based
on
the
ACT
data,
the
data
contained
in
the
industry
letters
to
OTC,
and
data
EPA
recently
collected,
there
is
considerable
agreement/
support
with
the
16.8
g/
bhp­
hr
uncontrolled
emission
rate
value
EPA
proposed.

Selective
Catalytic
Reduction
Information
received
by
EPA
from
the
natural
gas
transmission
industry
after
publication
of
the
NOx
SIP
Call
final
rule
in
1998
indicate
that
most,
if
not
all,
large
natural
gas­
fired
leanburn
IC
engines
in
the
SIP
Call
region
are
in
natural
gas
distribution
and
storage
service
and
that
these
engines
experience
frequently
changing
load
conditions.
According
to
the
industry,
these
conditions
make
application
of
SCR
infeasible.
The
industry
also
stated
that
low
emission
combustion
(
LEC)
technology
is
a
proven
technology
for
natural
gas­
fired
lean­
burn
engines,
while
SCR
is
not.
24
Regarding
variable
load
operations,
EPA's
ACT
document
states
that
little
data
exist
with
which
to
evaluate
application
of
SCR
for
the
lean
burn,
variable
load
operations.
More
recent
information
indicates
that
application
of
SCR
on
variable
load
engines
experienced
problems
in
earlier
applications
but
that
vendors
of
SCR
systems
believe
they
have
corrected
the
earlier
problems
with
a
new
generation
of
the
SCR
technology.
25
However,
SCR
still
remains
to
be
widely
demonstrated
in
the
United
States
on
lean
burn
IC
engines
in
variable
load
operation.
With
the
understanding
that
these
large
IC
engines
are
in
variable
load
operations,
EPA
believes
there
is
an
insufficient
basis
currently
to
conclude
that
SCR
is
an
appropriate
technology
for
the
large
variable
load
lean­
burn
engines.
Therefore,
EPA
no
longer
believes
that
SCR
is
a
highly
cost­
effective
control
technology
for
the
natural
gas­
fired
lean­
burn
IC
engines.

Emission
Rate
with
Low
Emission
Combustion
(
LEC)
Technology
The
industry
and
EPA
agree
that
low
emission
combustion
(
LEC)
technology
is
a
proven
technology
for
natural
gas­
fired
lean­
burn
engines.
26
The
ACT
for
IC
engines
and
other
15
Morgan
(
Natural
Gas
Pipeline
Company
of
America).

27From
Copper­
Bessemer,
a
reasonable
level
of
performance
expected
to
be
achieved
by
LEC
retrofits
is
3
g/
hp­
hr.
According
to
another
major
vendor
(
Dresser­
Rand/
Clark),
LEC
has
no
problem
meeting
the
3.0
g/
hp­
hr
level
even
for
Worthington
engines.
See
docket
at
XII­
E­
14
and
XII­
E­
15.
documents
also
indicate
that
LEC
technology
is
appropriate
for
lean­
burn
engines,
continuous
or
variable
load,
and
is
highly
cost
effective.
The
EPA
proposed
that
application
of
LEC
would
achieve
NO
x
emission
levels
in
the
range
of
1.5­
3.0
g/
bhp­
hr.
This
is
an
82­
91
percent
reduction
from
the
average
uncontrolled
emission
levels
reported
in
the
ACT
document
and
discussed
above.
IC
engine
manufacturers
will
typically
guarantee
the
LEC
performance
to
be
3.0
g/
bhp­
hr
or
less.
27
1.
Data
on
large
IC
engines
with
LEC
technology
In
2002
EPA
collected
additional
data
on
emission
rates
of
lean
burn
engines
that
have
been
retrofitted
with
LEC.
These
engines
had
been
identified
as
being
retrofitted
with
LEC
in
Interstate
Natural
Gas
Association
of
America's
(
INGAA)
April
22,
2002
comments
on
the
proposed
Phase
2
SIP
Call.
Also,
earlier
emission
test
results
had
been
obtained
for
several
engines
retrofitted
with
LEC
including
7
Clark
TLA­
6
(
2,000
HP)
engines
at
Southern
Cal
Gas's
Newberry
Springs
Station.
Three
emission
tests
were
performed
on
each
of
the
7
engines.
The
3­
test
averages
range
from
0.8
­
1.7
g/
bhp­
hr.
Also,
a
Cooper
Bessemer
GMV­
6
located
at
Kittanning,
New
York
was
retrofitted
with
LEC
and
tested
by
GRI.
The
3
emission
test
results
were
1.4,
1.8
and
2.5
g/
bhp­
hr
(
average
­
1.9
g/
bhp­
hr).
Also,
emission
test
data
were
obtained
from
several
state
agencies.
The
results
for
all
these
engines
are
summarized
in
Table
4.

Table
4.
Large
IC
Engines
Tested
with
Retrofit
LEC
Controls
Engine
Model
Number
of
engines
tested
Test
Results
(
g/
bhp­
hr)
%
of
total
units
in
the
SIP
Call
Area
Clark
BA­
8T
3
1.3,
3.1,
3.2
1
Clark
HLA
6
1.7,
1.9,2.4,
2.5,2.7,
2.8
(
Avg
­
2.3)
5
Clark
TLA
20
0.4­
4.0
(
others
­
0.5(
2)
0.8.
0.9(
2),
1.0,
1.1,
1.2,
1.3(
2),
1.4
(
2),
1.7,1.9,
2.3,
2.4(
2),
2.9)
Avg
­
1.5
1
16
28The
total
percentage
(
models
with
and
without
test
data;
80
and
21)
do
not
add
to
100
due
to
rounding
convention.
Clark
TCV
6
1.4­
3.6(
others
­
2.5,
3.0,
3.3,3.5)
Avg­
2.9
18
Cooper­
Bessemer(
C­
B)
GMW
2
0.7,
4.3
17
C­
B
V­
250
8
1.6
­
3.4(
2)
(
others
­
2.6,
2.8;
3.0,
3.2,
3.3)
Avg
­
2.9
12
C­
B
GMWA
1
0.6
8
C­
B
GMWC
3
3.1(
3
engines
tested)
6
C­
B
GMVA
2
0.5
,3.3
Avg
­
1.9
2
C­
B
12V­
275
2
1.3,
3.1
0
C­
B
8Q155L
1
1.9
0
C­
B
GMV
1
1.9
0
C­
B
W­
330
1
0.5
1
Ingersoll­
Rand(
I­
R)
KVG
4
Avg
­
2.0
1
I­
R
KVR
2
1.4,
2.1
1
I­
R
KVS
13
0.4,
1.1,
1.2,
1.3,
2.3,
2.5,
2.6,
2.8,
3.0,
3.0,
3.3,
3.6,
3.7
7
Totals
75
0.4
­
4.3
80
Models
without
test
data
­
C­
B
LSV
­
6%;
Worthington
MLV
­
3%;
Clark
TCVC
­
3%;
C­
B
Z­
330
­
2%;
C­
B
GMVH
­
2%;
Nordberg
FSE
­
1%;
I­
R
KVB
­
1%;
Worthington
­
1%;
C­
B
GMWH
­
1%;
C­
B
GMWS
­
1%
Total
­
21%
28
The
data
in
Table
4
show
that
56
of
75
engines
with
LEC
retrofits
have
NOx
emission
test
levels
that
are
at
or
below
3.0
g/
hp­
hr.
Nineteen
of
75
engines
(
25
%)
have
emission
test
results
17
29"
Stationary
Reciprocating
Internal
Combustion
Engines:
Updated
Information
on
NOx
Emissions
and
Control
Techniques,"
EC/
R
Incorporated,
September
1,
2000,
page
4­
5.

30
Docket
number
XII­
D­
24
greater
than
3.0
g/
hp­
hr
with
the
maximum
being
4.3
g/
hp­
hr.
The
next
highest
was
4.0
g/
hp­
hr.
The
average
emission
level
achieved
by
these
75
engines
is
2.2
g/
hp­
hr.

The
data
in
Table
5
below
use
the
same
data
as
in
Table
4,
except
the
data
are
limited
to
large
engines
in
the
NOx
SIP
call
area.
The
data
show
that
40
of
the
56
tests
have
NOx
emission
levels
at
or
below
3.0
g/
bhp­
hr.
The
LEC
technology
retrofit
on
these
large
engines
achieved,
on
average,
an
emission
rate
of
2.3
g/
bhp­
hr.

The
set
of
data
for
large
engines
in
the
SIP
Call
area
cover
80%
of
the
engine
models
in
the
NOx
SIP
call
area.
However,
emission
rates
for
some
of
the
engine
models
for
which
test
data
are
not
available
are
likely
to
be
higher
than
the
2.3
average
value.
For
example,
Worthington
and
Nordberg
engines
are
known
to
be
difficult
to
retrofit.
One
vendor
reported
achieving
a
level
of
6
g/
bhp­
hr
for
certain
Worthington
engines.
29
A
Worthington
UTC
165
in
New
York
reduced
NOx
emissions
to
4.4
g/
hp­
hr.
A
pipeline
company
commented
that
they
operate
6
Worthington
engines
and
that
4.0
g/
bhp­
hr
is
their
targeted
emission
reduction
level,
based
on
vendor
projections.
30
Thus,
it
appears
that
a
4.0
to
6.0
g/
bhp­
hr
level
is
achievable
on
these
difficult
to
retrofit
Worthington
engines.
At
this
time,
EPA
believes
that
5.0
g/
bhp­
hr
is
a
reasonable
emission
rate,
on
average,
for
engines
known
to
be
difficult
to
retrofit.
Although
not
all
of
the
21%
of
engine
models
for
which
test
data
are
not
available
are
likely
to
be
difficult
to
retrofit,
EPA
believes
it
is
reasonable
to
treat
these
engines
as
one
group
and
to
conservatively
assume
that
this
group
of
engines
would
achieve
a
5.0
level,
on
average.

Table
5.
Large
IC
Engines
in
SIP
Call
Area
Tested
with
Retrofit
LEC
Controls
Engine
Model
Number
of
engines
tested
Test
results
(
g/
hp­
hr)

Clark
BA­
8T
3
1.3,
3.1,
3.2
Clark
HLA
6
1.7,
1.9,
2.4,
2.5,2.7,
2.8
Clark
TCV
5
1.7;
3.0,3.3,
3.5,
3.6
Clark
TLA
13
0.4,
0.5,
0.5,
1.1,
1.3,
1.3,
1.4,
1.9,
2.3,
2.4,
2.4,
2.9,
4.0
C­
B
12V­
275
2
1.3,
3.1
18
31For
large
lean­
burn
IC
engines
in
the
NOx
SIP
Call
states,
2­
stroke
engines
represent
83%
of
the
total
large
engines
and
85%
of
the
total
large
engine
horsepower.
(
From
INGAA's
April
22,
2002
comments,
pages
2
and
10.)
C­
B
GMV
1
1.9
C­
B
GMW
2
0.7,
4.3
C­
B
GMWA
1
0.6
C­
B
GMWC
1
3.1
C­
B
V­
250
8
1.6,
2.6,
2.8,
3.0,
3.2,
3.3,
3.4,
3.4
C­
B
W­
330
1
0.5
Cooper
Quad
8Q155L
1
1.9
I­
R
KVS
12
1.1,
1.2,
1.3,
2.3,
2.5,
2.6
2.8,
3.0,
3.0,
3.3,
3.6,
3.7
Totals
56
0.4
­
4.0
(
Avg
­
2.3)

The
data
in
Tables
4
&
5
were
disaggregated
below
for
2­
and
4­
stroke
engines
(
Tables
6­
9
below).
In
Tables
6
and
7,
data
for
the
large
IC
engines
with
LEC
retrofit
indicate
controlled
levels
of
2.2
g/
bhp­
hr
for
both
2­
and
4­
stroke
engines.
Test
data
for
the
large
IC
engines
with
LEC
retrofit
in
the
SIP
call
area
indicate
controlled
levels
of
2.3
and
2.5,
respectively,
for
the
2­
and
4­
stroke
engines
(
Tables
8
and
9).
Assuming
85%
of
the
engines
in
the
SIP
Call
area
are
2­
stroke,
31
the
weighted
average
of
the
2.3
and
2.5
values
is
2.3.
Thus,
based
on
the
available
data,
the
emission
factor
is
the
same
whether
considering
2­
and
4­
stroke
engines
together
or
separately.

Table
6.
Large
IC
Engines
Tested
with
Retrofit
LEC
Controls
­­
2
stroke
Engine
Model
Number
of
engines
tested
Test
Results
(
g/
bhp­
hr)
%
of
total
units
in
the
SIP
Call
Area
Clark
BA­
8T
3
1.3,
3.1,
3.2
(
Avg
­
2.5)
1
Clark
HLA
6
1.7,
1.9,
2.4,
2.5,
2.7,
2.8
(
Avg­
2.3)
5
Clark
TCV
6
1.4­
3.6
(
other
tests
­
3.0,
3.3,3.5,
2.5)
Avg­
2.9
18
19
Clark
TLA
20
0.4­
4.0
(
others
­
0.9,
0.8,
0.9,
1.0,
1.2,
1.7,
0.5,0.5,1.4,
1.9,
2.3,2.4,2.4,1.4,1.3,1.1,
1.3,
2.9)
Avg
­
1.5
1
Cooper­
Bessemer
(
C­
B)
8Q155L
1
1.9
0
C­
B
12V­
275
2
1.3,
3.1
(
Avg
­
2.2)

C­
B
GMV
1
1.9
0
C­
B
GMVA
2
0.5,
3.3
Avg
­
1.9
2
CB
GMW
2
0.7,
4.3;
AVG
­
2.5
17
C­
B
GMWA
1
0.6
8
C­
B
GMWC
3
3.1
(
3
engines
tested)
6
C­
B
V­
250
8
1.6
­
3.4
(
other
­
2.8;
3.4,
3.3,3.0,2.6,3.2)
Avg
­
2.9
12
C­
B
W­
330
1
0.5
1
Total
56
engines
Avg
­
2.2
Table
7
B
Large
IC
Engines
Tested
with
Retrofit
LEC
Controls
­­
4
stroke
Ingersoll
Rand
(
I­
R)
KVG
4
Avg
­
2.0
1
I­
R
KVR
2
1.4
­
2.1(
Avg
­
1.8)
1
I­
R
KVS
13
0.4,
1.1,
1.2,
1.3,
2.3,
2.5,
2.6,
2.8,
3.0,
3.0,
3.3,
3.6,
3.7;
Avg
­
2.4
7
Total
19
Avg­
2.2
20
Table
8
­
Large
IC
Engines
in
SIP
Call
Area
Tested
with
Retrofit
LEC
Controls
­­
2
stroke
Engine
Model
Number
of
engines
tested
Test
results
(
g/
hp­
hr)

C­
B
12V­
275
2
1.3,
3.1
(
Avg
­
2.2)

C­
B
GMV
1
1.9
C­
B
GMW
2
0.7,
4.3
(
Avg­
2.50
C­
B
GMWA
1
0.6
C­
B
GMWC
1
3.1
C­
B
V­
250
8
1.6,2.6,2.8,3.0,3.2,3.3,
3.4,
3.4
(
Avg
­
2.9)

C­
B
W­
330
1
0.5
Cooper
Quad
8Q155L
1
1.9
Clark
BA­
8T
3
1.3,
3.1,
3.2
(
Avg
­
2.5)

Clark
HLA
6
1.7,
1.9,
2.4,
2.5,2.7,
2.8
(
Avg
­
2.3)

Clark
TCV
5
1.7;
3.0,3.3,
3.5,
3.6
(
Avg­
3.0)

Clark
TLA
13
0.4,
0.5,
0.5,
1.1,
1.3,
1.3,
1.4,
1.9,
2.3,
2.4,
2.4,
2.9,
4.0
(
Avg­
1.7)

Total
44
engines
Avg­
2.3
Table
9
­
Large
IC
Engines
in
SIP
Call
Area
Tested
with
Retrofit
LEC
Controls
­­
4
stroke
I­
R
KVS
12
1.1,
1.2,
1.3,
2.3,2.5,
2.6
2.8,
3.0,
3.0
.3.3,
3.6,
3.7
(
Avg
­
2.5)

As
shown
in
Table
10,
the
maximum
NOx
emission
level
for
the
13
engines
with
an
HPFI
retrofit
was
2.4
g/
hp­
hr.
The
average
was
1.1
g/
hp­
hr.
High­
pressure
fuel
injection
(
HPFI)
uses
high
pressure
fuel
injector
systems
to
enhance
the
mixing
of
air
and
fuel
in
the
combustion
cylinder.
According
to
a
control
equipment
vendor,
HPFI
does
not
require
precombustion
chambers
or
as
much
excess
air.
Reducing
the
amount
of
excess
air
required
would
diminish
the
21
32EC/
R
­
p.
4­
24.

33See
www.
enginuityinc.
com/
products/
HPFi.
htm.
turbocharging
and
intercooling
requirements.
HPFI
could
significantly
reduce
the
cost
and
complexity
of
retrofits.
HPFI
is
sometimes
used
in
LEC
retrofits
and
also
may
be
used
in
combination
with
ignition
timing
adjustment
and
improved
A/
F
ratio
and
ignition
system
controls.
32
According
to
another
HPFI
vendor,
HPFI
has
a
fraction
of
the
cost
of
traditional
combustion
retrofit
technology
and
reduces
NOx
by
up
to
80%;
CO
emissions
up
to
50%;
and
has
up
to
8%
fuel
savings.
33
Table
10
.
Large
IC
Engines
with
Retrofit
HPFI
Engine
Model
Number
of
engines
tested
Test
Results
(
g/
bhphr
%
of
total
units
in
the
SIP
Call
Area
C­
B
GMW
6
0.4,
0.5(
2);
0.6,
0.8,
1.0
17
C­
B
GMWA
4
0.7,
0.8,
0.9,
1.0
8
I­
R
KVS
3
2.1,
2.3,
2.4
7
Totals
13
0.4
­
2.4(
Avg­
1.1)
32
2.
Data
on
IC
Engines
with
LEC
that
are
not
large,
retrofit
gas
pipeline
engines
Data
on
the
performance
of
LEC
for
new
IC
engine
models
that
are
used
by
the
natural
gas
pipeline
industry
are
contained
in
the
ACT
and
other
documents.
These
results
are
shown
in
Table
11.
Seventeen
engines
with
test
results
were
reported
with
test
results
that
vary
from
1.0
­
6.0
g/
hp­
hr.
The
next
highest
test
result
was
2.6
g/
hp­
hr.
The
6.0
g/
hp­
hr
is
contained
in
the
ACT
which
considers
this
test
result
not
to
be
representative
of
the
achievable
controlled
NOx
emission
level
of
LEC.
The
average
of
all
data
including
the
6.0
is
1.8
g/
hp­
hr.

Table
11
­
NOx
Emissions
for
New
Large
IC
Engines
with
LEC
Engine
Model/
Location
Controlled
(
G/
hp­
hr)
Reference
Clark
TCV
­
10
(
2
engines)
2.6
ACT
(
p.
5­
68)

Clark
TCV­
10
1
GRI
Transmission
Report
22
Clark
TCVD(
2
engines)
1.6,
1.6
Sanders
Memo;
INGAA
­
9/
01;
p.
33
C­
B
GMVH
­
10,
12
6.0,
1.5
ACT
­
p.
5­
68
C­
B
GMVH
1.4
INGAA
­
2/
17/
99
C­
B
Q155HC/
Consumers
Energy/
Ray
Station/
MI
(
2
engines)
2.0,
2.0
6/
7/
02
email
­
Dennis
Dunlap
C­
B
W330/
Tn
Gas
Station
241
­
NY
0.6
5/
29/
02
Fax
from
Ted
Gardella
C­
B
W330/
Columbia
Gas
­
Crawford,
OH(
2
engines)
1.4,
1.4
6/
14/
02
email
from
John
Paslevicz
I­
R
KVS/
National
Fuel
Gas
Supply
1.0
5/
24/
02
email
­
Ted
Gardella
I­
R
KVSE
(
2,100
­
2,900
HP)
1.2
INGAA
(
9/
01­
­
p.
33);
Sanders
Memo
­
Ref.
4
I­
R
412
1.1
INGAA
2/
17/
99
­
Attachment
C
Superior
16SGTB/
Columbia
Gas
­
Gala
Station/
VA
1.1
Telecon
with
Dean
Down/
Roanoke,
Va
Data
from
rebuilt
engines
were
also
available.
These
results
are
shown
in
Table
12.
There
were
emission
test
results
on
ten
engines
whose
models
are
used
by
the
natural
gas
pipeline
industry.
These
results
vary
from
0.5
­
2.5
g/
hp­
hr
with
an
average
of
1.2
g/
hp­
hr.

Table
12
­
NO
x
Emissions
for
Large
IC
engines
rebuilt
with
LEC
Engine
Model
Controlled
NOx
(
g/
hp­
hr)
Reference
C­
B
10V­
250
1.3
ACT­
p.
5­
68
C­
B
GMV/
So
Cal
Gas
­
Goleta,
CA
­
1,100
HP
0.6
EC/
R
­
p.
4­
8;
INGAA
­
9/
01
­
p.
40
C­
B
GMVA­
8/
Mobil
­
Ventura
Co,
CA
3.0
EC/
R
­
p.
4­
6;
INGAA
­
9/
01
­
p.
30
C­
B
GMVA/
Santa
Barbara
Co,
CA
­
Engine
67
0.5
INGAA
9/
01
­
D­
2
23
C­
B
W330
­
PG&
E
­
Hinckley,
CA(
2
engines)
1.0,
1.3
EC/
R
­
p.
4­
8;
INGAA­
9/
01
­
F­
5
I­
R
KVS/
So
Cal
Gas
­
Aliso
Canyon,
CA
(
3
engines)
0.5,
0.6,
0.6
EC/
R
­
p.
4­
8
I­
R
KVS­
412
­
Williams
Gas
Pipeline
Station
505
­
NJ
2.5
INGAA
­
9/
01
­
p.
34
3.
Miscellaneous
LEC
Data
There
are
other
data
on
the
performance
of
LEC
on
engines
that
are
not
large
engines
(
that
is,
engines
that
emit
less
than
1
TPD
of
NO
x)
or
are
not
used
by
the
natural
gas
pipeline
industry
or
are
not
retrofit
LEC
installations.
The
data
listed
below
are
primarily
from
new
IC
engines
with
factory­
installed
LEC
technology.

The
ACT
on
Table
5­
5
(
p.
5­
38)
has
data
on
5
rich
burn
engines
that
were
retrofit
to
LEC
using
a
precombustion
chamber.
The
engines
range
in
size
from
1,200
to
2,000
HP.
Emission
test
results
range
from
0.37
­
2.0
g/
hp­
hr.
Table
5­
9
in
the
ACT
provides
information
on
LEC
used
on
4
lean
burn
IC
engines
(
3
rebuilt
and
1
new
engine):
NO
x
emissions
range
from
0.5
­
1.8
g/
hp­
hr
and
engine
sizes
range
from
4,000
to
7,000
HP.

In
the
EC/
R
report,
there
are
various
references
with
LEC
test
data.
Ventura
County,
California
has
320
tests
on
23
engines
on
8
engine
models.
Emissions
range
from
0.1
­
4.0
g/
bhp­
hr
with
an
average
of
0.7
g/
bhp­
hr.
Only
1
test
was
greater
than
3.0
g/
bhp­
hr.
From
Santa
Barbara
County,
California
there
were
12
tests
on
2
rebuilt
engines
and
1
new
engine.
The
engine
range
in
size
from
1,100
­
1,800
HP.
Emission
test
results
range
from
0.1­
0.7
g/
hp­
hr.
From
San
Diego
County,
California
there
were
121
tests
from
13
new
engines
of
5
engine
models.
Emission
test
results
range
from
0.3­
4.8
g/
hp­
hr.
The
average
test
result
was
1.1
g/
hp­
hr.
Only
1
of
the
121
emission
tests
was
above
3.0
g/
hp­
hr(
the
4.8
g/
hp­
hr).
Also
data
from
So
Cal
Gas's
­
Honor
Rancho
location
was
obtained
on
5
engines
that
are
each
5,500
HP.
There
were
7
tests
that
range
from
0.4
­
0.7
g/
bhp­
hr.
The
average
emissions
were
0.6
g/
bhp­
hr.

Also
emission
data
were
summarized
in
an
EPA
memo
dated
May
19,
2000.
(
Sanders
memo).
In
addition
to
the
test
data
already
mentioned,
7
engines
Santa
Barbara
County
that
range
in
size
from
25
­
410
HP.
There
were
a
total
of
24
emission
tests
for
these
engines
that
range
from
0.05
­
1.5
g/
hp­
hr.
A
1996
GRI
reference
in
this
memo
has
emission
test
data
on
4
engines
and
4
engine
models
that
range
in
size
from
1,800­
4,200
HP
that
are
used
by
the
gas
pipeline
industry.
The
19
emission
test
results
for
these
engines
range
from
0.3
­
3.1
g/
hp­
hr.

Data
supplied
by
INGAA
in
1999
to
EPA
are
also
summarized
in
this
memo.
There
are
18
emission
tests
on
4
new
IC
engines
from
3
engine
models.
Tests
results
range
from
0.7
­
3.1
g/
hp­
hr.
The
average
is
1.6
g/
hp­
hr.
24
34The
total
percentage
(
models
with
and
without
test
data;
80
and
21)
do
not
add
to
100
due
to
rounding
convention.
For
purposes
of
the
weighted
average
calculation
a
79/
21
split
is
The
Sanders
memo
also
cites
test
results
that
are
contained
in
the
1997
AP­
42.
There
were
a
total
of
15
emission
tests
for
2
and
4
stroke
engines.
For
2­
stroke
engines,
the
average
was
1.1
g/
hp­
hr
and
for
the
4­
stroke
engines­
0.6
g/
hp­
hr.
The
size
of
the
engine
is
uncertain
and
whether
the
engine
is
new,
retrofit
or
rebuilt.

Also,
test
data
on
a
rebuilt
Ingersoll­
Rand
KVS­
412
at
Transcontinental
Gas
Pipeline
Station
505
in
Neshanic,
NJ
is
presented.
For
this
2,050
HP
engine,
emissions
were
2.4
g/
hp­
hr
from
a
uncontrolled
estimate
of
21.5
g/
hp­
hr.

Emission
test
results
were
obtained
on
two
Texas
plants.
Transco's
Station
40
at
Sour
Lake,
Texas
has
a
Waukesha
3521
GL
­
600
hp
had
emission
test
results
of
0.61,
0.74,
0.68
g/
hp­
hr.

Colorado
Interstate
Gas
Station
at
Masterson,
TX
has
a
White­
Superior
8GTLX­
2­
825(
1,070
HP)
engine
which
had
a
lean
burn
conversion.
Four
emission
tests
results
were
0.5,
0.6,
0.9
and
1.8
g/
hp­
hr
or
an
average
of
1.0
g/
hp­
hr
From
www.
energyalliance.
com/
GMC/
GMC99/
monday/
ingersoll.
html,
the
first
ever
LEC
retrofit
of
I­
R
KVG
is
reported
at
Texas
Eastern
Transmission
­
6
engines
in
the
Beaumont­
Port
Arthur
area.
These
are
rich­
burn
engines
with
a
NOx
permit
limit
of
2.0
g/
bhp­
hr
and
CO
permit
limit
of
3.0
g/
bhp­
hr
across
the
engine's
normal
operating
range
75­
105%
rated
torque.
The
control
was
designed
by
Enginuity
and
consisted
of
static­
mixing
single
point
injection
system
and
water­
cooled
screw­
in
PCC
was
used
as
the
high
ignition
source.
No
modification
to
the
heads
was
required.

From
information
supplied
by
Sam
Clowney
to
OTC
on
11/
20/
00,
a
Worthington
UTC
165
in
NY
reduced
NOx
from12.0
­
4.4
g/
hp/
hr;
a
63%
reduction.

The
average
of
test
results
for
engines
with
LEC
that
are
not
large
or
not
used
by
the
natural
gas
pipeline
industry
or
are
not
retrofit
was
well
below
3.0
g/
hp­
hr.
Only
4
of
the
approximately
82
engines
exceed
3.0
g/
hp­
hr.
The
highest
reading
was
a
Worthington
engine
at
4.4
g/
hp­
hr
and
quite
a
few
engines
were
below
1.0
g/
hp­
hr.
These
data
generally
show
that
installation
of
LEC
technology
on
this
group
of
engines
produces
emissions
less
than
3.0
g/
bhphr
on
average.

3.
Summary:
emission
rates
with
LEC
technology
In
summary,
based
on
the
available
test
data,
EPA
believes
it
is
reasonable
to
assume
79%
of
the
large
engines
in
the
SIP
Call
area
are
able
to
meet
a
2.3
level,
on
average,
and
that
21%
are
able
to
meet
a
5.0
level,
on
average,
with
LEC
technology.
34
Thus,
calculating
the
weighted
25
used.
The
resultant
percentage
reduction
value,
83%,
is
the
same
if
the
split
is
79/
21
or
80/
20.

35EC/
R
report
on
IC
engines,
section
4.1.2.

36March
3,
1999
letter
from
J.
W.
Hibbard,
Cooper
Energy
Services,
to
Bill
Neuffer,
EPA;
March
4,
1999
telecon
summary
of
call
between
Joe
Hibbard,
Cooper
Energy
Services
and
Bill
Neuffer,
EPA;
and
letter
of
May
7,
1999
from
Charles
Wilke,
Dresser­
Rand
Company,
to
Bill
Neuffer.
average
for
installation
of
LEC
technology
retrofit
on
all
of
these
large
IC
engines
results
in
a
2.9
g/
bhp­
hr
emission
rate.

Availability
of
LEC
Technology
As
described
in
the
ACT
document,
LEC
technology
for
lean­
burn
IC
engines
generally
means
the
modification
of
a
natural
gas
fueled,
spark
ignited,
reciprocating
internal
combustion
engine
to
reduce
emissions
of
NO
X
by
utilizing
ultra­
lean
air­
fuel
ratios,
high
energy
ignition
systems
and/
or
pre­
combustion
chambers,
increased
turbocharging
or
adding
a
turbocharger,
and
increased
cooling
and/
or
adding
an
intercooler
or
aftercooler.
Because
there
are
many
types
of
existing
lean
burn
engines
(
e.
g.,
some
turbocharged,
some
not),
the
retrofit
of
LEC
technology
would
require
different
modifications
depending
on
the
particular
engine.

The
EPA
believes
that
LEC
retrofit
kits
are
available
for
virtually
all
affected
lean­
burn
engines.
This
is
based
on
the
EC/
R
report
on
IC
engines,
35
references
cited
in
the
9­
5­
00
TSD,
36
and
additional
information
described
below.
The
EPA
also
obtained
information
from
various
IC
engine
manufacturers.
This
information
is
summarized
in
Table
10
below.

Table
10
­­
Availability
of
Retrofit
LEC
for
Various
Large
IC
Engine
Models
in
SIP
Call
Area
Engine
Model
Number
of
Engines
%
of
Total
Units
%
of
Total
HP
LEC
Available?

Clark
TCV
28
18
22
Yes
Cooper­
Bessemer
(
C­
B)
GMW
26
17
10
Yes
C­
B
V­
250 
19
12
13
Yes
C­
B
GMWA
12
8
5
Yes
Ingersoll­
Rand(
I­
R)
KVS
11
7
4
Yes
C­
B
LSV
10
6
7
Yes
26
37Telecon
with
Ron
Billig
­
7/
12/
02;
docket
number
XII­
E­
14.
C­
B
GMWC
9
6
5
Yes
Clark
HLA
8
5
3
Yes
Worthington
MLV
5
3
4
Yes
Clark
TCVC
4
3
8
Yes
C­
B
Z­
330
3
2
6
Yes
C­
B
GMVH
3
2
1
Yes
C­
B
GMVA
3
2
1
Yes
Clark
TCVD
2
1
3
Yes
I­
R
KVR
2
1
2
Yes
Nordberg
FSE
2
1
1
??

Clark
TLA
2
1
1
Yes
C­
B
W­
330
1
1
1
Yes
I­
R
KVT
1
1
1
Yes
Clark
BA
1
1
0.3
Yes
I­
R
KVG
1
1
0.2
Yes
Worthington
ML
1
1
1
Yes
C­
B
GMWH
1
1
1
Yes
C­
B
GMWS
1
1
1
Yes
Total
156
100
100
All
but
2
of
156
engines
For
Cooper­
Bessemer
engines,
All
2
and
4
cycle
Cooper
engines
(
Cooper­
Bessemer,
Enterprise,
Superior,
Ajax)
can
be
retrofitted
with
LEC;
either
Clean
Burn
or
EcoJet.
Also
the
EcoJet
can
be
adapted
to
any
IC
engine
model
including
Worthingtons
and
Clarks.
The
Clean
Burn
system
can
only
be
installed
on
a
Cooper
engine
(
Cooper,
Enterprise,
Ajax,
Superior).
37
For
Clark,
Ingersoll­
Rand
engines
several
sources
of
information
were
obtained.
Low
cost
PCC
retrofits
are
available
for
engines
that
are
Clark
TLA,
TLAB­
D;
TCV,
TCVA­
27
38"
Low­
Cost
NOx
Controls
for
Pipeline
Engines"
See
docket
number
XII­
K­
93
or
www.
gastechnology.
org/
pub/
oldcontent/
pubs3/
trans/
tp_
lcncpe.
html
39Telecon
dated
6/
7/
02;
docket
number
XII­
E­
15.

40"
The
SIP
combustion
System
for
NOx
Reductions
on
Existing
Dresser­
Rand
Gas
Engines"
see
(
Docket
XII­
K­
96)
or
(
www.
dresser­
rand.
com/
e­
tech/
tp014/
tp014prt.
htm).

41"
Low­
Cost
Nox
Controls
for
Pipeline
Engines,"
see
docket
at
XII­
K­
90
or
www.
gastechnology.
org/
pub/
oldcontent/
pub3/
trans/
tp­
inger.
html,

42Pechan
IC
engines
report.
Annual
costs
in
1990
$
per
ozone
season
tons
reduced.
Note:
1990
$
are
used
in
order
to
easily
compare
with
the
NOx
SIP
call's
"
highly
cost
effective"
value
of
$
2000/
ton
(
in
1990
$).

43EC/
R
report
on
IC
engines,
section
2.2.
Annual
costs
in
1990
$
per
ozone
season
tons
reduced.
($
460­
910
in
1997
dollars).
D;
HLA,
BA,
HBA
models.
38
According
to
Dresser­
Rand
personnel,
the
screw­
in
prechamber
(
SIP)
has
been
installed
on
79
engines
at
7
different
owner/
operators
in
5
different
states.
The
SIP
can
be
installed
on
any
Dresser­
Rand,
Ingersoll­
Rand,
Clark
or
Worthington
engine.
39
Screw­
in
prechambers
are
available
for
TCV,
TCVA,
TVAD,
TLA,
TLAD,
TCVC,
LA,
HLA,
BA,
HBA,
RA,
HLA,
KVS,
KVS,
KVR,
and
KVT.
40
LEC
using
lean­
burn
operation,
precombustion
chambers,
and
enhanced
in­
cylinder
mixing
of
fuel
and
air
can
be
applied
to
Ingersoll­
Rand
KVS,
KVS,
KVT,
TVS,
KVR,
and
KVS
models
regardless
of
the
number
of
cylinders.
41
Cost
Effectiveness
of
LEC
Technology
The
average
cost
effectiveness
for
large
IC
engines
using
LEC
technology
was
estimated
in
the
Pechan
IC
engines
report
to
be
$
532/
ton
(
ozone
season).
42
The
EC/
R
report
on
IC
engines
estimates
the
average
cost
effectiveness
for
IC
engines
using
LEC
technology
to
range
from
$
420­
840/
ton
(
ozone
season)
for
engines
in
the
2,000­
8,000
bhp
range.
43
The
key
variables
in
determining
average
cost
effectiveness
for
LEC
technology
are
the
average
uncontrolled
emissions
at
the
existing
source,
the
projected
level
of
controlled
emissions,
annualized
costs
of
the
controls,
and
number
of
hours
of
operation
in
the
ozone
season.
The
ACT
document
uses
an
average
uncontrolled
level
of
16.8
g/
bhp­
hr,
a
controlled
level
of
2.0
g/
bhp­
hr
(
87%
decrease),
28
44EC/
R
report
on
IC
engines,
section
5.1.4.

45The
1993
ACT
document
for
IC
engines
uses
a
cost
of
$
2,440
for
annual
testing,
page
6­
5.
In
"
CAPCOA/
ARB
Proposed
Determination
of
Reasonably
Available
Control
Technology
and
Best
Available
Retrofit
Control
Technology
for
Stationary
Internal
Combustion
Engines,"
December
3,
1997
the
document
estimates
testing
costs
at
$
3,000
per
engine
(
pg.
52).

46Telephone
records
by
Bill
Neuffer
dated
5­
18­
00,
(
two)
5­
19­
00,
and
5­
24­
00.
and
nearly
continuous
operation
in
the
ozone
season.
The
EPA
believes
the
ACT
document
provides
a
reasonable
approach
to
calculating
cost
effectiveness
for
LEC
technology.

The
EPA
acknowledges
that
specific
values
will
vary
from
engine
to
engine.
For
additional
information,
we
have
included
sensitivity
analyses
in
this
TSD
regarding
the
key
variables
for
cost
effectiveness:
uncontrolled
and
controlled
levels,
hours
of
operation,
and
annualized
costs.
The
sensitivity
analyses
are
summarized
later
in
this
TSD
and
indicate
a
range
of
cost
effectiveness
for
large
IC
engines
using
LEC
technology
of
$
540­
890/
ton
(
ozone
season).

Other
Cost
and
Analysis
Factors
Monitoring
costs
In
the
NO
x
SIP
call
rulemaking,
EPA
assumed
continuous
emissions
monitoring
systems
(
CEMS)
might
be
required
by
States
that
chose
to
regulate
IC
engines.
The
EPA
now
believes
that
CEMS
may
not
be
necessary
unless
an
engine
is
participating
in
a
trading
program.
Alternate
monitoring
approaches,
such
as
parametric
monitoring
and/
or
annual
testing,
are
less
costly
and
may
be
sufficient
to
assure
compliance.
Monitoring
of
pressure,
which
may
be
correlated
with
temperature
and,
thus,
NOx
emissions,
is
a
form
or
parametric
monitoring
that
may
be
successfully
applied
at
a
cost
of
less
than
$
1000/
year.
44
Annual
testing
would
add
about
$
3,000/
year.
45
Time
to
Implement
Controls
for
IC
Engines
The
pipeline
industry
has
considerable
experience
with
the
installation
of
LEC
technology.
Based
on
information
primarily
from
manufacturers
of
control
equipment,
and
from
a
regulatory
agency
and
operator
of
the
control
equipment,
EPA
believes
the
time
between
a
request
for
cost
proposal
and
field
installation
on
a
few
engines
can
be
less
than
11
months.
46
However,
installing
controls
on
many
engines
in
a
narrow
time
frame
is
more
problematic.
As
discussed
below,
EPA
believes
that
a
reasonable
time
frame
is
24
months
from
the
SIP
submittal
date
and
that
the
initial
compliance
date
should
occur
within
the
ozone
season.

The
EPA
obtained
additional
information
regarding
this
issue.
One
manufacturer
29
47
See
docket
number
XII­
E­
01.

48See
docket
number
XII­
E­
02.

49See
http://
www.
dieselsupply.
com/
dscartic.
htm
for
reprint
of
article
from
May
1998
of
"
American
Oil
&
Gas
Reporter."

50August
22,
2002
memo
from
Lydia
Wegman
to
EPA
Regional
Air
Directors
providing
guidance
on
issues
related
to
stationary
internal
combustion
engines
and
the
NOx
SIP
call.
estimated
the
time
between
request
for
cost
proposal
and
contract
to
be
2­
5
months
and
typically
3­
4
months.
It
then
takes
4­
5
months
for
delivery
and
an
additional
1
month
to
install
and
commence
operation.
This
adds
up
to
a
total
of
7­
11
months.
47
Another
manufacturer
estimated
the
time
between
cost
proposal
and
contract
is
2­
4
weeks
to
obtain
bids;
2­
3
months
for
selection
of
bids;
12­
20
weeks
for
parts
delivery
to
site;
and
2
weeks
to
1
½
month
for
field
installation.
48
Another
manufacturer
estimated
from
request
for
cost
bids
to
shipping
of
parts
takes
6­
8
months
for
delivery
and
an
additional
2­
4
weeks
to
install
and
commence
operation.
This
adds
up
to
a
total
of
6
½
­
6
months.
17
Information
from
the
Ventura
County
Air
Pollution
Control
District
in
California
estimated
2
weeks
to
1
month
to
install
LEC
and
the
total
time
estimated
from
request
for
cost
proposal
and
commencing
operation
of
LEC
was
6­
9
months.
A
gas
pipeline
company,
CMS
Energy,
stated
that
a
compliance
schedule
of
11
months
was
easy
to
meet
for
1­
2
engines
but
would
put
a
stress
on
the
system
for
200
engines.
Columbia
Gas
Transmission
Corporation
installed
controls
on
2
engines
in
Bedford
Co.,
PA
in
three
days,
meeting
the
3.0
g/
bhp­
hr
standard
set
by
the
State.
49
Thus,
there
is
some
agreement
that
the
necessary
compliance
period
for
installation
of
controls
on
a
small
number
of
engines
is
less
than
one
year.

The
EPA
expects
some
companies
to
choose
to
phase­
in
installation
of
the
control
equipment
over
a
2­
year
period
(
or
longer
if
the
companies
begin
retrofit
activities
sooner)
and
that
installation
activities
would
occur
primarily
in
the
summer
along
with
normally
scheduled
maintenance
activities.
Further,
as
noted
below,
not
all
of
the
potentially
affected
IC
engines
should
be
expected
to
need
LEC
retrofits
and
not
in
the
same
time
frame.

In
response
to
Phase
II
of
the
NOx
SIP
call,
some
States
may
seek
emission
reductions
from
source
categories
other
than
IC
engines.
Other
States
have
already
met
their
NOx
budgets
and
do
not
need
to
further
control
IC
engines
for
purposes
of
the
NOx
SIP
call.
Still
other
States
have
met
at
least
a
portion
of
the
Phase
II
NOx
SIP
Call
reductions
due
to
emission
reductions
affecting
other
source
categories
contained
in
their
1­
hour
ozone
nonattainment
area
plans.
This
reduces
the
need
to
retrofit
IC
engines
in
those
States.

In
many
cases,
companies
may
use
"
early
reductions"
achieved
at
IC
engines
due
to
other
requirements,
such
as
RACT.
50
For
example,
many
IC
engines
were
previously
controlled
to
meet
RACT
requirements
in
many
of
the
NOx
SIP
call
States.
These
emission
reductions
help
States
meet
their
NOx
budgets
and,
thus,
decrease
the
amount
of
additional
reductions
needed.
30
51"
IC
Engine
OTAG
Questions"
document
prepared
by
INGAA,
2/
17/
00.
Many
of
these
engines
are
smaller
than
the
"
large"
engines
identified
in
the
NOx
SIP
Call.

52Alpha
Gamma
memo
of
6­
19­
02.

53INGAA
letter
of
July
16,
2002.

54A
top­
end
overhaul
is
generally
recommended
between
8,000
and
30,000
hours
of
operation
that
entails
a
cylinder
head
and
turbocharger
rebuild
(
see
Table
4
from
"
Technology
Characterization:
Reciprocating
Engines"
prepared
by
Energy
Nexus
Group
for
EPA,
2­
02).

55
GRI
12­
98
report
"
NOx
Control
for
Two­
Cycle
Pipeline
Reciprocating
Engines,"
page
4­
11.
According
to
a
information
submitted
by
INGAA,
a
1996­
97
survey
determined
that
245
lean
burn
engines
in
the
SIP
Call
area
have
LEC.
51
Many
engines
in
the
NOx
SIP
call
area
already
have
decreased
NOx
emissions
at
rich­
burn
engines
through
NSCR.
52
States
may
choose
to
credit
these
reductions
instead
of
requiring
new
reductions
at
other
engines
in
order
to
meet
the
SIP
budget.
Many
more
NOx
reductions
are
likely
to
result
from
future
MACT
controls
at
IC
engines.
These
factors
also
reduce
the
need
to
retrofit
IC
engines
in
some
States.

Some
pipeline
companies
will
phase­
in
the
control
equipment
over
a
multi­
year
time
frame.
53
Stretching
out
the
installation
time
frame
in
this
manner
would
help
the
companies
achieve
the
results
on
time.
Further,
companies
might
choose
to
install
controls
early
in
some
of
their
engines
in
a
time
frame
that
coincides
with
the
engine
rebuild
cycle.
54
In
another
case,
installation
of
the
LEC
retrofit
kit
was
estimated
to
span
3
to
4
weeks
and
the
installation
was
not
expected
to
impact
the
normal
maintenance
interval.
55
These
approaches
will
help
reduce
the
time
needed
to
install
the
controls.

The
EPA
believes
the
industry
has
demonstrated
that
multiple
engines
at
compressor
stations
can
be
successfully
retrofit
over
a
24
month
time
frame.
For
example,
the
Jefferson
Town
Compressor
Station's
RACT
compliance
plan
of
April
2000
describes
the
installation
of
LEC
using
a
phased
approach
over
a
2
yr
period.
Four
engines
were
retrofit
during
summer
2001
and
the
remaining
5
engines
were
retrofit
in
summer
2002.
Each
engine
was
expected
to
be
out
of
service
for
approximately
6
weeks
and,
due
to
heavy
demand
during
winter
heating
season,
all
engines
were
expected
to
be
operable
from
October
­
April.
Two
additional
cases
show
installation
on
multiple
engines
in
short
time
periods.
Southern
California
Gas
Co.
completed
testing
of
one
engine
in
1995
and
installed
precombustion
chambers
on
six
engines
in
its
Mojave
Desert
operating
area.
The
conversion
of
the
first
unit
was
completed
in
October
1995
and
the
conversion
of
the
sixth
unit
was
in
November
1996.
The
engines
met
the
2.0
g/
bhp­
hr
standard
set
by
the
Mojave
Air
District.
Furthermore,
as
cited
in
a
case
study
in
Vidor,
Texas,
6
engines
in
31
56
See
http://
www.
enginuityinc.
com
57
"
Determination
of
RACT
and
BARCT
for
Stationary
Spark­
Ignited
Internal
Combustion
Engines,"
California
Air
Resources
Board,
November
2001,
pg.
IV­
15.

58August
22,
2002
memo
from
Lydia
Wegman
to
EPA
Regional
Air
Directors
providing
guidance
on
issues
related
to
stationary
internal
combustion
engines
and
the
NOx
SIP
call.
the
Beaumont/
Port
Arthur
area
were
retrofitted
in
summer
of
1999.56
As
shown
below,
EPA
also
examined
historic
time
frames
allowed
by
the
Congress
and
various
regulatory
agencies
to
achieve
compliance
with
NOx
requirements
following
State/
local
rule
adoption.
These
time
frames
generally
illustrate
the
successful
implementation
of
past
regulatory
programs
involving
the
installation
of
NOx
controls.

In
the
1990
amendments
to
the
CAA,
Congress
added
RACT
requirements
for
major
sources
of
NOx.
All
categories
of
major
NOx
sources
in
certain
areas
of
the
nation
were
required
to
install
RACT
as
expeditiously
as
practicable
or
no
later
than
May
31,
1995.
Thus,
Congress
allowed
a
maximum
of
30
months
from
the
SIP
submittal
deadline
of
November
15,
1992
for
a
much
larger
number
of
sources
than
affected
by
this
rulemaking.

Subsequent
to
the
initial
set
of
NOx
RACT
SIP
revisions,
EPA
approved
NOx
RACT
SIP
submittals
in
some
areas
which
had
been
exempt
from
the
requirements.
For
example,
in
Dallas,
SIP
rules
required
RACT
as
expeditiously
as
practicable
or
24
months
from
the
State
adoption
date
(
rule
adopted
March
21,
1999).
The
State
of
Texas,
on
December
31,
1997,
implemented
a
requirement
for
all
major
NO
x
sources
in
the
Houston
area
to
implement
RACT;
the
State
adopted
a
compliance
date
of
November
15,
1999
for
this
program
(
22.5
months).
In
a
recent
case,
the
State
of
Louisiana
allowed
up
to
a
3­
year
period
in
Baton
Rouge,
coinciding
with
their
attainment
deadline.

For
engines
subject
to
RACT
limits,
the
California
Air
Resources
Board
guidance
document
on
IC
engines
recommends
final
compliance
within
two
years
of
district
rule
adoption.
57
The
guidance
states
that
this
time
period
should
be
sufficient
to
evaluate
control
options,
place
purchase
orders,
install
equipment,
and
perform
compliance
verification
testing.
The
Sacramento
Air
District
in
California
required
compliance
within
2
years
of
rule
adoption
(
June
1995).

Furthermore,
EPA
believes
that
States
will
process
permits
expeditiously,
especially
those
permits
associated
with
pollution
control
projects.
The
EPA
has
specifically
encouraged
States
in
a
recent
memo
to
consider
exempting
pollution
control
projects
from
certain
permitting
requirements.
58
Further,
by
moving
the
compliance
date
to
at
least
24
months
after
the
SIP
submittal
date,
EPA
believes
that
the
time
needed
to
revise
permits
will
not
adversely
affect
the
compliance
schedule.
32
In
summary,
several
factors
described
above
will
serve
to
minimize
the
number
of
large
IC
engines
that
would
need
to
be
scheduled
for
LEC
retrofit.
Further,
companies
that
phase­
in
compliance
activities
over
several
years
would
also
reduce
the
number
of
IC
engines
needing
LEC
retrofit
per
year.
It
is
important
to
note
that
RACT
experience
shows
that
companies
can
install
LEC
retrofit
over
a
2­
year
time
frame,
even
where
multiple
engines
are
located
at
the
same
compressor
station.
In
recent
RACT
compliance
time
decisions,
State/
Local
regulatory
agencies
generally
specified
24
month
periods
to
install
controls.
The
Congress
in
its
1990
CAA
amendments
allowed
a
maximum
of
30
months
for
all
major
NOx
sources
across
the
nation
to
install
RACT;
this
was
a
much
larger
task
than
installation
of
controls
at
IC
engines
in
certain
States.
As
a
result,
EPA
believes
that
a
2­
year
period
after
the
SIP
submittal
due
date
is
adequate
for
the
installation
of
controls.

In
addition,
because
the
NOx
SIP
call
is
directed
at
emissions
during
the
ozone
season,
EPA
believes
that
the
initial
month
where
compliance
is
required
should
occur
during
the
ozone
season.
Therefore,
the
compliance
date
is
24
months
from
the
SIP
submittal
date
if
the
SIP
submittal
date
occurs
during
the
ozone
season
or,
if
not,
24
months
from
the
SIP
submittal
date
plus
the
days
until
the
next
ozone
season
begins
(
May
1).

Increased
Power
Output
and
Fuel
Savings
Implementation
of
LEC
may
yield
additional
benefits
of
fuel
economy
and
power
output.
Up
to
5%
fuel
economy
improvement
is
reported
in
the
ACT
document
from
installing
LEC
(
p.
7­
12).
The
1990
GRI
report
describes
"
cost
credit
due
to
improved
engine
performance"
and
states
that
fuel
economy
can
be
improved
up
to15%
and
power
output
65%
(
p.
10).
The
1994
GRI
report
indicates
increases
in
power
output
but
slight
losses
in
fuel
economy
associated
with
controls
that
achieve
80­
90%
NO
x
reduction.
The
1996
AP­
42
indicates
improved
power
output
and
fuel
efficiency
with
LEC
(
sect.
3.2.4.2).
In
the
Pechan
IC
engines
report
cited
earlier
in
this
TSD,
a
1%
fuel
savings
is
included
in
the
cost
analysis.

A
CARB
report
"
Sources
and
Control
of
Oxides
of
Nitrogen
Emissions"
­
August
1997
states
that
at
the
80
%
reduction
level,
the
efficiency
of
the
precombustion
chamber
is
often
improved
over
that
of
an
uncontrolled
engine.
At
reductions
of
more
than
90
%
which
is
obtained
by
carefully
controlling
operating
parameters
and
extreme
leaning
of
the
air/
fuel
mixtures,
there
is
usually
some
decrease
in
engine
efficiency.

Types
of
IC
Engines
In
the
February
22,
2002
proposed
rule,
EPA
invited
comment
on
how
many
of
the
large
natural
gas­
fired
IC
engines
are
from
lean­
burn
operation
and
how
many
from
rich­
burn.
The
INGAA
commented
that
156
of
the
168
large
engines
listed
in
the
NOx
SIP
Call
Inventory
that
have
SIC
codes
associated
with
the
natural
gas
transmission
industry
are
lean­
burn
models,
with
one
exception.
According
to
INGAA,
the
other
12
engines
are
no
longer
in
service,
are
owned
by
33
59INGAA
document
dated
9/
01,
page
A­
8.

60Pechan
IC
engines
report.
a
company
not
included
in
the
industry
data
base
or
are
duplicates.
All
but
one
engine
is
lean
burn
and
the
majority
are
2­
stroke
engines.
59
For
the
purposes
of
calculating
the
IC
engine
portion
of
the
NOx
SIP
Call
state
budgets,
INGAA
recommended
that
EPA
should
assume
that
all
the
large
natural
gas
fired
stationary
engines
in
the
inventory
are
lean
burn.
Thus,
the
vast
majority
of
large
IC
engines
in
the
NOx
SIP
call
inventory
are
natural
gas­
fired
lean­
burn
engines.
Furthermore,
the
emission
inventory
does
not
contain
sufficient
detail
to
determine
exactly
which
engines
are
lean
burn
and
which
are
not.
For
these
reasons
EPA
agrees
with
the
comment
that
it
is
reasonable
to
assume
that
all
the
large
natural
gas
stationary
engines
in
the
inventory
are
lean­
burn
for
the
purposes
of
calculating
the
IC
engine
portion
of
the
NOx
SIP
Call
state
budgets.

Results
of
Cost
and
Sensitivity
Analyses
The
discussion
below
summarizes
an
August
11,
2000
report
by
the
Pechan­
Avanti
Group
which
estimated
the
control
costs
and
NO
x
emission
reductions
for
large
IC
engines
affected
under
the
NO
x
SIP
Call.
The
report
provides
information
about
the
universe
of
potentially
affected
stationary
IC
engines,
control
cost
modeling
methods,
scenario
analyses,
and
caveats
and
uncertainties
associated
with
this
analysis.
60
The
results
of
the
analyses
are
summarized
below.
Additional
information
is
contained
in
EPA's
9­
5­
00
TSD.

The
average
cost
per
ton
(
ozone
season)
for
the
main
analysis
is
$
532
per
ton.
This
ozone
season
cost
per
ton
is
affected
mostly
by
the
natural
gas­
fired
engine
control
costs.
The
uncontrolled
NO
x
emission
level
is
16.8
g/
bhp­
hr.
For
purposes
of
this
analysis,
the
controlled
NO
x
level
with
LEC
is
2.0
g/
bhp­
hr.
Oil­
fired
engines
are
about
3
percent
of
the
population
of
large
IC
engines.
While
oil­
fired
engine
costs
are
just
above
$
1,000
per
ton,
they
have
a
negligible
influence
on
regionwide
costs.

In
Scenario
B,
the
control
efficiency
for
low
emission
combustion
applied
to
lean
burn
natural
gas­
fired
engines
is
reduced
to
82
percent
(
3
g/
hp­
hr).
This
increases
the
average
cost
per
ton
by
$
20/
per
ton.
The
tons
of
NO
x
decreased
by
about
2,000
tons
in
the
ozone
season,
compared
to
the
87%
reduction
in
the
main
analysis.

Scenario
C
increases
the
NO
x
control
efficiency
for
lean
burn
engines
to
90
percent.
This
additional
emission
reduction
reduces
the
average
cost
per
ton
to
about
$
520
per
ton,
which
is
$
12
per
ton
less
than
in
the
main
analysis.
34
Scenario
D
changes
the
uncontrolled
NO
x
emission
level
for
lean
burn
gas­
fired
engines
to
13.7
g/
bhp­
hr
from
16.8
g/
bhp­
hr.
With
fewer
NO
x
tons
being
reduced,
this
raises
the
cost
per
ton
to
$
603
per
ton.

A
control
level
of
1.2
g/
bhp­
hr
(
93%
decrease)
in
Scenario
E
produces
the
lowest
average
cost
per
ton
of
$
513
(
and
the
largest
emission
reduction).

Scenario
F
reduces
annual
operating
hours
to
6,500.
This
changes
both
the
emission
reductions
and
the
costs.
Compared
with
other
scenarios,
there
are
fewer
emission
reductions
but
lower
costs,
resulting
in
a
cost
per
ton
$
49
higher
than
the
main
analysis.

Scenario
G
retains
the
capital
cost
estimates
that
were
used
in
the
September
1998
Non­
Electricity
Generating
Unit
(
EGU)
cost
analysis
for
the
NO
x
SIP
Call.
The
scenario
has
the
same
emission
reductions
as
the
main
analysis,
but
with
$
334
per
ton
higher
estimated
costs.
