1
MEMORANDUM
DATE:
January
5,
2004
SUBJECT:
Emission
Limitations
for
the
Stationary
Reciprocating
Internal
Combustion
Engines
Final
Rule
FROM:
Tanya
Ali
Alpha­
Gamma
Technologies,
Inc.

TO:
Sims
Roy,
EPA
OAQPS
ESD
Combustion
Group
The
purpose
of
this
memorandum
is
to
document
the
emission
limitations
that
have
been
established
for
the
final
national
emission
standards
for
hazardous
air
pollutants
(
NESHAP)
for
stationary
reciprocating
internal
combustion
engines
(
RICE).

Background
On
December
19,
2002,
EPA
proposed
NESHAP
for
stationary
RICE
(
67
FR
77830).
The
emission
limitations
EPA
proposed
are
presented
in
Table
1.

Table
1.
Proposed
Emission
Limitations
for
Stationary
RICE
Subcategory
Emission
Limitation
Existing,
new,
and
reconstructed
4
stroke
rich
burn
(
4SRB)
stationary
RICE
Reduce
formaldehyde
emissions
by
75
percent
or
more,
if
you
use
non­
selective
catalytic
reduction
(
NSCR);

or
limit
the
concentration
of
formaldehyde
to
350
parts
per
billion
by
volume
dry
basis
(
ppbvd)
or
less
at
15
percent
oxygen
(
O
2),
if
you
do
not
use
NSCR.
Subcategory
Emission
Limitation
2
New
and
reconstructed
2
stroke
lean
burn
(
2SLB)
stationary
RICE
Reduce
carbon
monoxide
(
CO)
emissions
by
60
percent
or
more,
if
you
use
an
oxidation
catalyst;

or
limit
the
concentration
of
formaldehyde
to
17
parts
per
million
by
volume
dry
basis
(
ppmvd)
or
less
at
15
percent
O
2,
if
you
do
not
use
an
oxidation
catalyst.

New
and
reconstructed
4
stroke
lean
burn
(
4SLB)
stationary
RICE
Reduce
CO
emissions
by
93
percent
or
more,
if
you
use
an
oxidation
catalyst;

or
limit
the
concentration
of
formaldehyde
to
14
ppmvd
or
less
at
15
percent
O
2,
if
you
do
not
use
an
oxidation
catalyst.

New
and
reconstructed
compression
ignition
(
CI)
stationary
RICE
Reduce
CO
emissions
by
70
percent
or
more,
if
you
use
an
oxidation
catalyst;

or
limit
the
concentration
of
formaldehyde
to
580
ppbvd
or
less
at
15
percent
O
2,
if
you
do
not
use
an
oxidation
catalyst.

Documentation
describing
EPA's
rationale
for
selection
of
maximum
achievable
control
technology
(
MACT)
floors
was
previously
submitted
to
the
docket
and
is
entitled
"
Existing
and
New
MACT
Floors
for
Stationary
Reciprocating
Internal
Combustion
Engines."
This
memorandum
can
be
obtained
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0067
(
A­
95­
35
II­
B­
44).

Based
on
comments
received
during
the
public
comment
period,
EPA
has
made
some
adjustments
to
the
emission
limitations
established
at
proposal.
However,
the
majority
of
emission
limitations
remain
unchanged.
The
final
rule
includes
adjusted
emission
limitations
for
existing,
new,
and
reconstructed
4SRB
stationary
RICE
and
for
new
and
reconstructed
2SLB
stationary
RICE.
The
following
sections
explain
EPA's
rationale
for
adjusting
the
emission
limitations
for
existing,
new,
and
reconstructed
4SRB
and
new
and
reconstructed
2SLB
stationary
RICE.
3
4
Final
Emission
Limitations
As
established
in
section
112(
d)
of
the
Clean
Air
Act,
MACT
is
the
maximum
degree
of
reduction
in
emissions
that
is
achievable,
considering
the
cost
of
achieving
the
reduction
and
the
health,
environmental,
and
energy
impacts.
The
MACT
standard
must
be
no
less
stringent
than
the
MACT
floor,
which
for
existing
sources
is
the
average
emission
limitation
achieved
by
the
best
performing
12
percent
of
existing
sources.
For
new
sources,
MACT
floor
is
the
emission
control
achieved
in
practice
by
the
best
controlled
similar
source.
To
be
a
similar
source,
a
source
should
not
have
any
characteristics
that
differ
sufficiently
to
have
a
material
effect
on
the
feasibility
of
emission
controls,
but
the
source
need
not
be
in
the
same
source
category
or
subcategory.

Existing,
New,
and
Reconstructed
4SRB
To
establish
the
emission
limitations
for
4SRB
engines
at
proposal,
EPA
found
the
lowest
percent
reduction
value
of
the
two
sources
tested.
However,
EPA
determined
that
27
percent
of
existing
4SRB
engines
use
NSCR
for
the
reduction
of
hazardous
air
pollutants
and
found
that
it
would
be
appropriate
to
use
only
the
higher
of
the
two
values
to
determine
the
emission
limitation
for
existing
4SRB
engines.
In
essence,
this
treated
the
top
performer
as
a
surrogate
for
the
top
half
of
the
population
using
NSCR,
or
the
top
13.5
percent
of
the
population
of
4SRB
engines.
The
EPA
followed
this
approach
at
proposal
since
it
is
more
closely
analogous
to
the
level
of
the
top
12
percent
of
sources
than
is
a
straight
average
of
the
two
sources
tested.

However,
in
reviewing
EPA's
approach
used
at
proposal,
EPA
feels
that
it
would
be
more
appropriate
to
include
in
the
analysis
the
data
from
the
lower
performing
of
the
two
engines,
thus
using
more
than
a
single
data
point
in
determining
the
emission
limitation
for
existing
engines.
Because
the
test
calculation
for
the
MACT
floor
for
existing
engines
is
supposed
to
be
based
on
the
average
of
the
top
performing
12
percent
of
sources,
it
would
be
better
to
rely
on
a
formula
that
does
not
rely
solely
on
the
highest
performer.
Also,
it
would
not
be
appropriate
to
use
a
straight
average
between
the
two
sources
tested,
because
that
would
not
be
a
fair
approximation
of
the
average
of
the
top
12
percent
of
sources.
Instead,
it
would
approximate
the
average
of
the
best
performing
27
percent
of
sources.
Therefore,
EPA
feels
a
reasonable
approach
is
to
discount
the
lower
performing
source
by
12/
27,
thus
reducing
the
influence
of
that
data
point
by
the
ratio
of
controlled
sources
(
27
percent
of
the
population
of
4SRB
engines)
compared
to
the
statutory
level
(
12
percent).
This
leads
to
a
weighted
average
where
the
data
point
for
the
lower
performing
engine
will
be
worth
22
percent
(
50
percent)
(
12/
27)
and
the
level
for
the
higher
performer
will
be
worth
78
percent.

In
addition,
to
be
consistent
with
the
approach
followed
for
other
engines
types,
i.
e.,
to
1The
calculation
of
percentage
reduction
is
as
follows:
(
lowest
tested
percentage
reduction
of
the
lower
performing
engine)
*
(.
222)
+
(
lowest
tested
percentage
reduction
of
the
higher
performing
engine)
*
(.
778)
=
(
75.5)(.
222)
+
(
76.2)(.
778)
=
76.0.
The
calculation
of
parts
per
billion
is
as
follows:
(
highest
tested
parts
per
billion
of
the
lower
performing
engine)
*
(.
222)
+
(
highest
tested
parts
per
billion
of
the
higher
performing
engine)
*
(.
778)
=
(
355)(.
222)
+
(
348)(.
778)
=
350.

5
establish
emission
limitations
based
on
test
results
from
tests
conducted
at
high
loads,
EPA
found
it
appropriate
to
exclude
runs
conducted
at
low
loads
in
determining
the
lower
and
higher
performer.
This
leads
to
a
final
emission
limitation
of
76
percent
control
efficiency
for
formaldehyde
or
350
ppbvd1
for
existing
4SRB
engines.

For
new
and
reconstructed
4SRB
engines,
EPA
proposed
a
formaldehyde
reduction
requirement
of
75
percent
and
an
alternative
formaldehyde
concentration
emission
limitation
of
350
ppbvd.
In
reviewing
the
4SRB
emissions
data
from
the
two
sources
tested
used
to
set
the
standard,
EPA
observed
that
the
minimum
percent
efficiency
achieved
by
the
best
performing
engine
was
actually
76.2
percent
formaldehyde
reduction.
Therefore,
the
EPA
acknowledges
that
the
proposed
formaldehyde
reduction
efficiency
should
have
been
set
at
76
percent
reduction
instead
of
75
percent
reduction.
The
EPA
has
adjusted
the
formaldehyde
reduction
requirement
for
new
and
reconstructed
4SRB
engines
to
76
percent
reduction
in
the
final
rule.

New
and
Reconstructed
2SLB
Comments
received
during
the
public
comment
period
stated
that
EPA
should
adjust
the
CO
percent
reduction
standard
for
new
and
reconstructed
2SLB
stationary
RICE
to
reflect
the
lowest
percent
reduction
achieved
during
EPA's
testing
at
Colorado
State
University
(
CSU).
The
EPA
has
adjusted
the
emission
limitation
for
new
and
reconstructed
2SLB
engines
from
60
percent
CO
reduction
to
58
CO
percent
reduction
to
fully
reflect
the
testing
at
CSU.
In
addition,
the
approach
EPA
followed
when
establishing
standards
for
other
engine
types
was
to
base
the
standards
on
emissions
data
obtained
from
runs
conducted
at
high
loads.
High
load
is
defined
as
100
percent
±
10
percent.
The
17
ppmvd
formaldehyde
concentration
was
based
on
a
run
conducted
a
low
load
(
69
percent).
The
EPA
therefore
found
it
inconsistent
to
use
the
data
from
the
low
load
run
to
establish
the
alternative
emission
limitation
for
new
and
reconstructed
2SLB
engines.
Instead,
EPA
has
adjusted
the
standard
and
has
specified
an
alternative
emission
limitation
of
12
ppmvd.
The
adjusted
standard
represents
the
highest
concentration
observed
at
CSU
at
high
loads.

Conclusion
As
discussed
in
this
memorandum,
EPA
has
adjusted
the
emission
limitation
for
existing,
new,
and
reconstructed
4SRB
engines
from
75
percent
formaldehyde
6
reduction
to
76
percent
formaldehyde
reduction
in
the
final
rule.
The
EPA
has
also
adjusted
the
emission
limitations
for
new
and
reconstructed
2SLB
engines
from
60
percent
CO
reduction
to
58
percent
CO
reduction,
and
from
17
ppmvd
formaldehyde
concentration
to
12
ppmvd
formaldehyde
concentration.
The
formaldehyde
concentration
emission
limitation
for
existing,
new,
and
reconstructed
4SRB
engines
and
the
emission
limitations
for
new
and
reconstructed
4SLB
engines
and
for
new
and
reconstructed
CI
engines
remain
unchanged.

In
the
final
rule,
EPA
has
specified
that
sources
can
comply
with
either
emission
limitation
option,
irrespective
of
control
technique
applied.
Engines
equipped
with
oxidation
catalyst
control
or
NSCR
may
comply
with
either
the
percent
reduction
emission
limitation
or
with
the
formaldehyde
concentration
emission
limitation.
Similarly,
engines
not
equipped
with
oxidation
catalyst
control
or
NSCR
may
choose
to
comply
with
either
the
percent
reduction
requirement
or
with
the
formaldehyde
concentration
requirement.
The
emission
limitations
EPA
has
specified
in
the
final
rule
are
presented
in
Table
2.

Table
2.
Final
Emission
Limitations
for
Stationary
RICE
Subcategory
Emission
Limitation
Existing,
new,
and
reconstructed
4SRB
stationary
RICE
Reduce
formaldehyde
emissions
by
76
percent1
or
more;

or
limit
the
concentration
of
formaldehyde
to
350
ppbvd
at
15
percent
O
2.

New
and
reconstructed
2SLB
stationary
RICE
Reduce
CO
emissions
by
58
percent
or
more;

or
limit
the
concentration
of
formaldehyde
to
12
ppmvd2
or
less
at
15
percent
O
2.

New
and
reconstructed
4SLB
stationary
RICE
Reduce
CO
emissions
by
93
percent
or
more;

or
limit
the
concentration
of
formaldehyde
to
14
ppmvd
or
less
at
15
percent
O
2.
Subcategory
Emission
Limitation
7
New
and
reconstructed
CI
stationary
RICE
Reduce
CO
emissions
by
70
percent
or
more;

or
limit
the
concentration
of
formaldehyde
to
580
ppbvd
or
less
at
15
percent
O
2.

1
If
the
source
commenced
construction
or
reconstruction
between
December
19,
2002
and
the
effective
date
of
the
final
rule,
it
may
reduce
formaldehyde
emissions
by
75
percent
or
more
until
3
years
after
the
effective
date
of
the
final
rule.

2
If
the
source
commenced
construction
or
reconstruction
between
December
19,
2002
and
the
effective
date
of
the
final
rule,
it
may
limit
the
concentration
of
formaldehyde
to
17
ppmvd
or
less
at
15
percent
O
2
until
3
years
after
the
effective
date
of
the
final
rule.
