1
MEMORANDUM
Subject:
Response
to
Public
Comments
on
Proposed
Stationary
Reciprocating
Internal
Combustion
Engines
NESHAP
From:
Sims
Roy,
ESD
Combustion
Group
To:
Robert
Wayland,
ESD
Combustion
Group
On
December
19,
2002,
the
EPA
proposed
national
emission
standards
for
hazardous
air
pollutants
(
NESHAP)
for
stationary
reciprocating
internal
combustion
engines
(
RICE)
(
67
FR
77830).
The
proposed
rule
fulfills
the
requirements
of
section
112(
d)
of
the
Clean
Air
Act
(
CAA).
The
purpose
of
this
document
is
to
present
a
summary
of
the
public
comments
that
EPA
received
on
the
proposed
standards
and
the
responses
developed
by
EPA.
This
summary
of
comments
and
responses
serves
as
the
basis
for
revisions
made
to
the
standards
between
proposal
and
promulgation.

The
EPA
received
64
public
comments
on
the
proposed
rule.
A
listing
of
all
persons
submitting
comments,
their
affiliation,
and
the
document
number
for
their
comments
is
presented
in
Table
1.
The
comments
can
be
obtained
online
from
the
EPA's
Edocket
website
(
http://
www.
epa.
gov/
edocket).
The
docket
number
for
this
rulemaking
is
OAR­
2002­
0059.
Each
commenter
has
two
document
numbers,
one
number
is
the
Edocket
Document
ID,
and
the
other
document
number
is
for
the
legacy
document
numbering
system.
In
this
document,
commenters
are
identified
by
the
last
three
digits
of
the
Edocket
Document
ID
of
their
comments.

On
January
21,
2003,
a
public
hearing
on
the
proposed
NESHAP
for
stationary
RICE
was
held
at
the
EPA
facility
in
Research
Triangle
Park,
NC.
Five
speakers
presented
comments
on
the
proposed
rule
and
are
listed
at
the
end
of
Table
1.
All
of
the
commenters
except
one
supplemented
their
comments
at
the
public
hearing
with
written
comments
submitted
to
the
docket.
Any
comments
from
the
public
hearing
that
were
not
included
in
the
written
comments
submitted
to
the
docket
are
also
summarized
in
this
document.
A
transcript
of
the
public
hearing
2
is
available
from
the
Edocket
website
(
document
ID
OAR­
2002­
0059­
0603).

Table
1.
List
of
Commenters
on
the
Proposed
NESHAP
for
Stationary
RICE
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
OAR­
2002­
0059­
0499
IV­
D­
02
L.
Eagan,
Chair,
STAPPA
Air
Toxics
Committee
and
R.
Colby,
Chair,
ALAPCO
Air
Toxics
Committee,
Washington,
DC.
01­
21­
03
OAR­
2002­
0059­
0500
IV­
D­
01
J.
Bardi,
Administrative
Assistant,
ASTM
International,
W.
Conshohocken,
PA.
12­
27­
02
OAR­
2002­
0059­
0526
IV­
D­
04
G.
M.
Adams,
Assistant
Departmental
Engineer
Office
Engineering
Department,
Los
Angeles
County
Sanitation
Districts
(
LACSD),
Whittier,
CA.
02­
12­
03
OAR­
2002­
0059­
0527
IV­
D­
45
H.
Malm
Ph.
D.,
Chief
Technical
Officer,
REM
Technology
Inc.,
Calgary
Alberta,
Canada.
02­
14­
03
OAR­
2002­
0059­
0528
jeanpublic@
yahoo.
com
undated
OAR­
2002­
0059­
0529
IV­
D­
46
S.
Kaderly,
Air
Quality
Division
Administrator,
Nebraska
Department
of
Environmental
Quality,
Lincoln,
NE.
02­
18­
03
OAR­
2002­
0059­
0530
IV­
D­
12
J.
P.
LaCosse,
President
&
Principal
Scientist,
Spectral
Insights
LLC,
Durham,
NC.
02­
17­
03
OAR­
2002­
0059­
0533
IV­
D­
10
M.
Heseding,
The
European
Association
of
Internal
Combustion
Engine
Manufacturers
(
EUROMOT),
Frankfurt,
Germany.
02­
18­
03
OAR­
2002­
0059­
0536
IV­
D­
18
J.
Shefchek,
Managing
Director
and
Chief
Environmental
Officer,
Alliant
Energy
Corp.,
Madison,
WI.
02­
13­
03
OAR­
2002­
0059­
0537
IV­
D­
08
J.
C.
Weisbrod,
JCW
Technologies,
Stevens
Point,
WI.
02­
17­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
3
OAR­
2002­
0059­
0540
IV­
D­
15
E.
M.
Adamo,
North
America
Environmental,
Air
Products
and
Chemical
Inc.,
Allentown,
PA.
02­
18­
03
OAR­
2002­
0059­
0542
IV­
D­
56
M.
W.
Stroben,
EHS
Manager,
Corporate
Environment
Health
&
Safety,
Duke
Energy,
Charlotte,
NC.
02­
19­
03
OAR­
2002­
0059­
0544
IV­
D­
47
J.
C.
deRuyter,
Principal
Consultant,
DuPont
Engineering
Technology,
Wilmington,
DE.
02­
20­
03
OAR­
2002­
0059­
0545
IV­
D­
48
D.
R.
Schregardus,
Deputy
Assistant
Secretary
of
the
Navy,
Washington,
DC.
02­
19­
03
OAR­
2002­
0059­
0546
IV­
D­
44
D.
E.
Donohoue,
Chief,
Emissions
Assessment
Branch
Stationary
Source
Division,
Air
Resources
Board,
Sacramento,
CA.
02­
20­
03
OAR­
2002­
0059­
0547
IV­
D­
24
J.
B.
Frank,
Counsel
to
the
Class
of
`
85
Regulatory
Response
Group,
Baker
Botts
LLP,
Washington,
DC.
02­
20­
03
OAR­
2002­
0059­
0548
IV­
D­
49
G.
Calvo,
Counsel
for
the
Utility
Air
Regulatory
Group,
Hunton
&
Williams,
Washington,
DC.
02­
20­
03
OAR­
2002­
0059­
0549
IV­
D­
54
T.
Steichen,
Sr.
Regulatory
Analyst,
American
Petroleum
Institute,
Washington,
DC.
02­
20­
03
OAR­
2002­
0059­
0550
IV­
D­
55
K.
King,
Supervisor,
Regulatory
&
Compliance
Support
Unit,
Air
Pollution
Control
Division,
State
of
Colorado,
Denver,
CO.
02­
14­
03
OAR­
2002­
0059­
0551
IV­
D­
31
D.
F.
Hunter,
Manager
Regulatory
Issues,
ConocoPhillips,
Houston,
TX.
02­
20­
03
OAR­
2002­
0059­
0552
IV­
D­
50
J.
Suchecki,
Director
Public
Affairs,
Engine
Manufacturers
Association,
Chicago,
IL.
02­
20­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
4
OAR­
2002­
0059­
0553
IV­
D­
23
J.
Michael
Geers,
Cinergy
Corp.,
Cincinnati,
OH.
02­
20­
03
OAR­
2002­
0059­
0554
IV­
D­
51
Hanover
Compression
Limited
Partnership,
Houston,
TX.
02­
20­
03
OAR­
2002­
0059­
0555
IV­
D­
32
T.
R.
Weeks,
Chief,
Engineering,
San
Diego
County
Air
Pollution
Control
District,
San
Diego,
CA.
02­
20­
03
OAR­
2002­
0059­
0556
IV­
D­
52
J.
Dreyer,
Director
of
Industry
Affairs,
Gas
Processors
Association,
Tulsa,
OK.
02­
20­
03
OAR­
2002­
0059­
0557
IV­
D­
20
J.
C.
Biechman,
Vice
President
Government
Affairs,
The
National
Fire
Protection
Association
(
NFPA),
Arlington,
VA.
02­
19­
03
OAR­
2002­
0059­
0558
IV­
D­
17
G.
A.
Wilkins,
Manager
Environmental
Support,
Marathon
Ashland
Petroleum
LLC,
Findlay,
OH.
02­
14­
03
OAR­
2002­
0059­
0559
IV­
D­
33
D.
L.
Kraisinger,
Vice
President,
Health,
Safety
&
Environment,
BP
America
Inc.,
Los
Angeles,
CA.
02­
20­
03
OAR­
2002­
0059­
0560
IV­
G­
01
T.
Pugh,
Manager
Environmental
Services,
American
Public
Power
Association
(
APPA),
Washington,
DC.
02­
24­
03
OAR­
2002­
0059­
0561
IV­
D­
41
P.
Faggert,
Vice
President
and
Chief
Environmental
Officer,
Dominion
Transmission
Inc.,
Glen
Allen,
VA.
02­
19­
03
OAR­
2002­
0059­
0562
IV­
D­
34
R.
J.
Barkanic,
Manager­
Environmental,
PPL
Services
Corp.,
Allentown,
PA.
02­
13­
03
OAR­
2002­
0059­
0564
IV­
D­
35
R.
J.
Morehouse,
ExxonMobil
Refining
&
Supply
Company,
Houston,
TX.
02­
19­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
5
OAR­
2002­
0059­
0565
IV­
D­
36
J.
J.
Mayhew,
Vice
President
Regulatory
&
Technical
Center
of
Expertise,
American
Chemistry
Council,
Arlington,
VA.
02­
20­
03
OAR­
2002­
0059­
0566
IV­
D­
13
J.
R.
Carson,
Senior
Staff
Engineer
Environmental
Affairs,
ISPAT
Inland
Inc.,
East
Chicago,
IN.
02­
17­
03
OAR­
2002­
0059­
0567
IV­
D­
14
J.
T.
Higgins,
Director
Bureau
of
Stationary
Sources
Division
of
Air
Resources,
New
York
State
Department
of
Environmental
Conservation,
Albany,
NY.
02­
18­
03
OAR­
2002­
0059­
0569
IV­
D­
16
R.
W.
Schenker,
Manager­
Air
Pollution
Control,
Corporate
Environmental
Programs,
General
Electric
Company,
Fairfield,
CT.
02­
18­
03
OAR­
2002­
0059­
0570
IV­
D­
38
J.
Jacobsen,
General
Counsel,
Alyeska
Pipeline,
Fairbanks,
AK.
02­
18­
03
OAR­
2002­
0059­
0571
IV­
D­
19
R.
W.
Stachowicz,
Senior
Development
Engineer,
Waukesha
Engine
Dresser,
Inc.,
Waukesha,
WI.
02­
18­
03
OAR­
2002­
0059­
0572
IV­
D­
21
J.
A.
Paul,
Supervisor,
Regional
Air
Pollution
Control
Agency,
Dayton,
OH.
02­
11­
03
OAR­
2002­
0059­
0573
IV­
D­
03
O.
M.
Dominguez,
Director,
Environmental
Management
Division,
National
Aeronautics
and
Space
Administration
(
NASA),
Washington,
DC.
02­
06­
03
OAR­
2002­
0059­
0574
IV­
D­
06
L.
Eagan,
Director,
Bureau
of
Air
Management,
State
of
Wisconsin
Department
of
Natural
Resources,
Madison,
WI.
01­
07­
03
OAR­
2002­
0059­
0575
IV­
D­
07
A.
G.
Berwick,
Associate
Director,
The
Clean
Energy
Group,
Concord,
MA.
02­
13­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
6
OAR­
2002­
0059­
0576
IV­
D­
05
D.
Samet,
Associate
Director,
Standards
Interpretation
Group,
Joint
Commission
on
Accreditation
of
Healthcare
Organizations
(
JCAHO),
Oakbrook
Terrace,
IL.
02­
13­
03
OAR­
2002­
0059­
0577
IV­
D­
09
B.
H.
McCully,
President,
American
Society
for
Healthcare
Engineering,
Chicago,
IL.
02­
18­
03
OAR­
2002­
0059­
0578
IV­
D­
39
N.
Popa,
Senior
Environmental
Planner,
Consumers
Energy,
Jackson,
MI.
02­
12­
03
OAR­
2002­
0059­
0579
IV­
D­
11
C.
Hornback,
Director,
Regulatory
Affairs,
Association
of
Metropolitan
Sewerage
Agencies,
Washington,
DC.
02­
18­
03
OAR­
2002­
0059­
0580
IV­
D­
40
P.
T.
Cavanaugh,
Vice
President
Federal
and
International
Government
Relations,
ChevronTexaco,
Washington,
DC.
02­
20­
03
OAR­
2002­
0059­
0581
IV­
D­
27
R.
Bessette,
President,
Council
of
Industrial
Boiler
Owners.
02­
19­
03
OAR­
2002­
0059­
0582
IV­
D­
30
V.
Patton,
Senior
Attorney,
Environmental
Defense,
Boulder,
CO.
02­
20­
03
OAR­
2002­
0059­
0583
IV­
D­
28
P.
A.
Lacey,
Senior
Managing
Counsel,
American
Gas
Association,
Washington,
DC.
02­
20­
03
OAR­
2002­
0059­
0584
IV­
D­
29
E.
W.
Repa,
Director
Environmental
Programs,
National
Solid
Wastes
Management
Assoc.,
Washington,
DC.
02­
13­
03
OAR­
2002­
0059­
0585
IV­
D­
22
N.
Dee,
Director
of
Environment
and
Safety,
National
Petrochemical
&
Refiners
Assoc.,
Washington,
DC.
02­
19­
03
OAR­
2002­
0059­
0586
IV­
D­
42
W.
O'Sullivan,
Acting
Director,
New
Jersey
Department
of
Environmental
Protection
(
NJDEP),
Trenton,
NJ.
02­
14­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
7
OAR­
2002­
0059­
0587
IV­
D­
37
L.
Beal,
Director
Environmental
Affairs,
Interstate
Natural
Gas
Association
of
America,
Washington,
DC.
02­
19­
03
OAR­
2002­
0059­
0588
IV­
G­
02
D.
Olson,
Administrator
Air
Quality
Division,
The
State
of
Wyoming
Department
of
Environmental
Quality,
Cheyenne,
WY.
02­
24­
03
OAR­
2002­
0059­
0593
IV­
D­
25
R.
D.
Langford,
Regulatory
Affairs
Manager,
Celanese
Acetate
LLC,
Narrows,
VA.
02­
20­
03
OAR­
2002­
0059­
0594
IV­
D­
26
J.
W.
Sanderson,
For
the
Colorado
Utilities
Coalition
for
Clean
Air
(
CUC),
Denver,
CO.
02­
20­
03
OAR­
2002­
0059­
0595
IV­
G­
03
R.
Meyer,
Manager
of
Environmental
Affairs,
American
Municipal
Power
­
Ohio
Inc.,
Columbus,
OH.
02­
27­
03
OAR­
2002­
0059­
0598
IV­
G­
04
M.
Murray,
Director,
Environmental
&
Safety
Policy,
Sempra
Energy,
San
Diego,
CA.
02­
28­
03
OAR­
2002­
0059­
0599
IV­
G­
05
D.
L.
Chapman,
Manager
Global
Environmental
Services,
The
Goodyear
Tire
&
Rubber
Company,
Akron,
OH.
03­
06­
03
OAR­
2002­
0059­
0600
IV­
G­
06
A.
Lawrence,
Director
Office
of
Environmental
Policy
and
Guidance,
Department
of
Energy,
Washington,
DC.
03­
18­
03
OAR­
2002­
0059­
0604
R.
L.
Greene.
04­
21­
03
OAR­
2002­
0059­
0605
IV­
D­
43
J.
Whynot,
Planning
and
Rules
Manager,
Planning,
Rule
Development
and
Area
Sources,
South
Coast
Air
Quality
Management
District,
Diamond
Bar,
CA.
02­
20­
03
OAR­
2002­
0059­
0606
R.
Claybrook,
Miratech
Corporation.
08­
07­
03
Edocket
ID
Number
Legacy
Document
Number
Commenter,
Addressee,
Title
or
Description,
etc.
Date
of
Document
8
OAR­
2002­
0059­
0603
(
transcript
of
public
hearing)
Speakers
were:
(
A)
B.
Nicholson,
Chief
of
the
Planning
Section,
NC
Division
of
Air
Quality,
on
behalf
of
the
State
and
Territorial
Air
Pollution
Program
Administrators
and
the
Association
of
Local
Air
Pollution
Control
Officials.

(
B)
S.
Clowney,
El
Paso
Pipeline
Group,
on
behalf
of
the
Interstate
Natural
Gas
Association
of
America.

(
C)
C.
Wait,
Panhandle
Pipe
Line
Companies,
on
behalf
of
the
Interstate
Natural
Gas
Association
of
America.

(
D)
M.
Fisher,
Wartsila
North
America.

(
E)
J.
Sucheki,
Engine
Manufacturers
Association.

Many
commenters
expressed
support
for
the
comments
submitted
by
other
commenters.
Table
2
shows
those
commenters
and
the
comments
that
they
supported.

Table
2.
List
of
Commenters
Expressing
Support
for
Other
Comments
Commenter
Supports
Comments
of:

OAR­
2002­
0059­
0542
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0544
OAR­
2002­
0059­
0565
OAR­
2002­
0059­
0551
OAR­
2002­
0059­
0549,
OAR­
2002­
0059­
0585,
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0554
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0556
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0559
OAR­
2002­
0059­
0565,
OAR­
2002­
0059­
0549
Commenter
Supports
Comments
of:

9
OAR­
2002­
0059­
0561
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0564
OAR­
2002­
0059­
0549,
OAR­
2002­
0059­
0565
OAR­
2002­
0059­
0567
OAR­
2002­
0059­
0499
OAR­
2002­
0059­
0578
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0580
OAR­
2002­
0059­
0549,
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0581
OAR­
2002­
0059­
0544
OAR­
2002­
0059­
0583
OAR­
2002­
0059­
0587
OAR­
2002­
0059­
0595
Agrees
with
OAR­
2002­
0059­
0594'
s
comments
in
commending
the
EPA
for
seeking
comment
on
various
aspects
of
CAA
section
112(
d)(
4).
Defers
to
OAR­
2002­
0059­
0594
on
the
use
of
RfDs.

The
summary
of
public
comments
and
responses
is
organized
as
follows:

I.
Applicability
A.
Affected
Source
B.
Applicability
Cutoff
C.
Exemptions
D.
Emergency/
Limited
Use
E.
Compression
Ignition
F.
Subcategories
G.
Other
II.
Definitions
A.
Rich/
Lean
Burn
B.
New/
Reconstructed
C.
Affected
Source
D.
Other
III.
Dates
A.
Comment
Period
B.
New/
Reconstructed
C.
Initial
Notification
D.
Initial
Compliance
E.
Other
IV.
Emission
Limitations
10
A.
Level
B.
Flexibility/
Options
C.
Compression
Ignition/
DPF
D.
Other
V.
Monitoring,
Recordkeeping,
Reporting
A.
CO
CEMS
B.
Operating
Parameters
1.
All
Subcategories
2.
4SRB
C.
Initial
Notification
D.
Startup,
Shutdown,
Malfunction
Plan
E.
Reports
F.
Other
VI.
Testing
A.
Test
Methods
1.
CARB
430
2.
FTIR
3.
ASTM
D6348
4.
EPA
Method
323
5.
EPA
SW­
846
Method
0011
6.
Other
B.
Performance
Tests
1.
Portable
CO
Monitor
2.
Other
VII.
Cost/
Impact
VIII.
Risk
IX.
Emissions
Data
A.
CSU
Testing
B.
Other
X.
Other
A.
General
Provisions
B.
Other
11
I.
APPLICABILITY
A.
Affected
Source
I.
A.
1
Comment:
One
commenter
(
569)
expressed
support
for
EPA's
decision
to
limit
the
RICE
maximum
achievable
control
technology
(
MACT)
to
major
sources.

Response:
The
EPA
has
noted
the
commenter's
support.
The
EPA
will
cover
area
sources
and
small
engines
under
CAA
sections
112(
k)
and
112(
d),
respectively,
in
a
later
rule.

I.
A.
2
Comment:
Two
commenters
(
564,
565)
requested
clarification
in
the
preamble
that
engines
less
than
or
equal
to
500
horsepower
(
HP)
are
included
in
the
RICE
affected
source.
One
commenter
(
564)
asked
that
EPA
clarify
that
emergency/
limited
use
engines,
and
existing
2
stroke
lean
burn
(
2SLB),
4
stroke
lean
burn
(
4SLB),
and
compression
ignition
(
CI)
engines
are
part
of
the
affected
source.
Commenter
564
stated
that
the
regulatory
language
(
§
63.6590(
b))
appears
to
indicate
that
engines
less
than
or
equal
to
500
HP,
emergency/
limited
use
engines,
and
existing
2SLB,
4SLB,
and
CI
RICE
are
part
of
the
affected
source
but
have
exceptions
from
rule
requirements.
However,
the
preamble
language
in
II.
A
(
67
FR
77833)
says
"
this
proposed
rule
applies
to
you
if...,
except
if
your
stationary
RICE
are
all
rated
at
or
under
500
brake
horsepower."
The
preamble
language
would
suggest
that
RICE
less
than
or
equal
to
500
HP
are
not
subject
to
the
rule
and
therefore
are
not
in
the
affected
source.

Response:
Engines
less
than
or
equal
to
500
HP
are
included
in
the
RICE
major
source
category,
but
are
not
covered
by
the
final
rule.
The
EPA
is
covering
area
sources
and
small
engines
in
a
later
rule.
The
EPA
recognizes
that
in
the
proposed
rule
it
was
unclear
which
engines
were
affected
sources
and
has
therefore
made
the
appropriate
revisions
to
the
regulatory
text
in
§
63.6590
to
clearly
indicate
that
the
final
rule
applies
to
stationary
RICE
with
a
site­
rating
of
more
than
500
HP
located
at
major
sources.
Therefore,
emergency
power
units,
limited
use
units,
engines
with
annual
landfill
and
digester
gas
consumption
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis,
and
existing
2SLB,
existing
4SLB,
and
existing
CI
RICE
are
affected
sources.
Existing
emergency
power
units,
existing
limited
use
units,
existing
engines
that
combust
landfill
gas
or
digester
gas
equivalent
to
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis,
existing
2SLB,
existing
4SLB,
and
existing
CI
engines
have
no
requirements
in
the
final
rule.
New
and
reconstructed
emergency
power
units
and
new
and
reconstructed
limited
use
units
must
only
submit
an
initial
notification.
New
and
reconstructed
engines
that
combust
landfill
gas
or
digester
gas
equivalent
to
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis
must
record
and
report
the
annual
fuel
usage
in
addition
to
submitting
an
initial
notification.

I.
A.
3
Comment:
One
commenter
(
545)
asked
that
EPA
clarify
§
§
63.6590(
b),
63.6600(
c),
and
63.6640(
f).
These
sections
are
missing
the
word
"
existing."

Response:
The
EPA
agrees
with
the
commenter
and
has
added
the
word
"
existing"
in
front
of
"
compression
ignition"
in
§
§
63.6590(
b)(
3)
(
previously
§
63.6590(
b)(
2)),
63.6600(
c),
and
12
63.6640(
e)
(
previously
§
63.6640(
f)).
The
EPA
has
also
added
"
existing"
in
front
of
"
4SLB"
in
§
63.6600(
c).

I.
A.
4
Comment:
One
commenter
(
570)
noted
that
§
63.6590(
b)(
2)
states
that
existing
2SLB,
4SLB,
and
CI
do
"
not
have
to
meet
the
requirements
of
this
subpart
and
of
subpart
A
of
this
part."
The
commenter
stated
that
§
63.6600(
c)
leaves
the
impression,
however,
that
some
requirements
for
these
sources
beyond
those
listed
in
Tables
1(
a),
2(
a),
1(
b),
and
2(
b)
of
the
proposed
rule
do
indeed
apply.
Similarly
with
§
63.6640(
f).
The
commenter
recommended
removing
§
63.6600(
c)
and
modifying
§
63.6640(
f)
to
eliminate
confusion.

Response:
The
EPA
feels
these
sections
are
clear
as
written
and
do
not
need
revision.
The
EPA
thinks
the
revisions
suggested
would
actually
add
confusion
because
it
would
omit
important
clarifying
information.

I.
A.
5
Comment:
One
commenter
(
585)
requested
clarification
on
what
is
considered
an
existing
RICE
unit
for
purposes
of
compliance.
According
to
the
commenter,
using
a
date
as
a
determination
whether
an
engine
is
existing
is
confusing.
The
commenter
stated
that
an
engine
takes
on
its
identity
when
first
assembled
into
an
engine
or
when
modified
to
be
a
different
kind
of
engine,
regardless
of
where
that
engine
is
ultimately
installed
or
whether
it
is
a
spare
on
the
shelf
awaiting
installation.
Another
commenter
(
565)
asked
that
EPA
clarify
that
an
existing
RICE
unit
is
any
engine
that
was
assembled
as
a
final
unit
before
December
19,
2002,
regardless
of
whether
it
was
or
has
been
installed
in
a
stationary
location.

One
commenter
(
565)
stated
that
the
criteria
that
makes
a
RICE
unit
affected
by
the
proposed
rule
does
not
limit
the
rule's
effects
to
only
units
that
operate.
The
proposed
factors
that
determine
applicability
are
construction
date,
manufacturer's
nameplate
rating,
and
specific
inherent
designs
of
units.
None
of
these
criteria
as
applied
in
the
proposal
include
a
requirement
that
the
engine
be
operational.
It
is
not
uncommon
for
an
owner
or
operator
to
have
idle
engines.
Some
may
be
installed
and
not
in
use.
Others
may
be
stored
for
later
use
as
replacements
or
spare
engines.
Importantly,
idle
units
are
distinct
from
emergency
power/
limited
use
units
because
an
idle
unit
is
not
in
any
use.
The
commenter
expressed
that
an
idle
RICE
unit
should
have
no
compliance
obligations
imposed
by
the
final
RICE
rule.

Response:
The
EPA
disagrees
with
the
first
set
of
comments
and
feels
that
the
date
an
engine
was
constructed
is
the
date
it
was
installed
at
the
operator
site
and
not
when
it
was
assembled
as
a
final
unit
at
the
manufacturer.
Thus,
any
engine
constructed
(
i.
e.,
installed
at
the
site
of
the
operator)
prior
to
December
19,
2002,
is
an
existing
engine
for
purposes
of
the
final
rule,
while
any
engine
constructed
on
or
after
that
date
is
a
new
engine.
For
purposes
of
the
final
rule,
the
term
"
on­
site
fabrication"
in
the
definition
of
construction
in
40
CFR
§
63.2
shall
refer
to
the
final
installation
at
the
site
of
the
final
operator.
This
definition
of
construction
is
in
line
with
how
EPA
generally
defines
construction,
i.
e.,
it
is
defined
by
when
the
unit
is
installed
at
the
operator's
location,
rather
than
where
it
is
first
assembled.
13
The
EPA
feels
it
is
appropriate
to
define
"
on­
site
fabrication"
as
the
final
site
of
installation
because
even
after
a
unit
has
been
manufactured,
several
components
necessary
in
order
to
be
able
to
operate
the
unit
must
be
considered
and
added.
The
owner
or
operator
cannot
go
directly
from
purchasing
the
unit
from
the
manufacturer
to
operation.
The
owner
or
operator
must
typically
have
a
building
to
house
the
unit
in,
construct
a
pad
for
the
unit,
run
utilities,
install
fuel
supply
tanks
or
run
the
natural
gas
line,
have
the
catalyst
vendor
install
the
pollution
control
equipment,
and
finally
test
the
unit
on­
site.
For
larger
engines
(
e.
g.,
5,000
HP
or
greater),
the
installation
process
is
even
more
pronounced.
For
these
reasons,
EPA
finds
it
appropriate
that
the
date
that
final
installation
of
the
unit
at
the
site
of
operation
is
commenced
should
be
considered
the
construction
date.

Engines
manufactured
prior
to
December
19,
2002,
but
where
installation
was
not
commenced
until
after
that
date,
are
considered
new
engines
and
must
comply
with
the
requirements
for
new
engines.
The
EPA
expects
that
these
units
will
be
able
to
comply
with
the
requirements
especially
since
the
control
equipment
is
typically
installed
on
the
engine
at
the
site
of
operation
and
does
not
come
with
the
engine
purchased
from
the
manufacturer.
Finally,
no
problems
are
expected
to
occur
with
retrofit
controls
because
the
control
technology
is
relatively
easy
to
retrofit,
especially
in
units
that
are
being
installed
initially
at
a
site.
If
owners
or
operators
anticipate
problems,
they
can
elect
to
purchase
a
new
engine
meeting
the
requirements
if
it
is
installed
after
that
date.

With
regard
to
the
next
comment,
the
EPA
disagrees
with
the
commenter's
proposition
that
EPA
needs
to
have
a
special
provision
to
deal
with
engines
that
are
installed
but
not
in
use.
For
new
engines
covered
by
the
final
rule,
which
will
be
the
vast
majority
of
the
engines,
the
final
rule
does
not
apply
until
startup
of
the
engine,
which
is
when
the
engine
begins
operation.
Therefore,
new
engines
are
not
covered
until
they
are
operational,
which
already
accomplishes
the
goal
of
the
commenter.
For
existing
engines,
EPA
feels
that
any
engine
that
does
not
meet
the
definition
of
limited
use
engine,
which
includes
engines
that
operate
less
than
100
hours
per
year,
should
not
be
relieved
of
compliance
obligations.
The
EPA
has
written
its
definitions
to
distinguish
emergency
engines
from
limited
use
engines,
which
should
reduce
some
confusion.
An
engine
that
does
not
operate
at
all
is
clearly
a
limited
use
engine,
which
by
definition
includes
engines
that
operate
0
hours
per
year.

I.
A.
6
Comment:
One
commenter
(
569)
asked
that
EPA
clarify
the
overlap
between
RICE
subcategories.
For
example,
it
is
not
clear
whether
an
emergency
power
unit
with
a
nameplate
rating
of
400
HP
is
subject
to
the
requirements
of
§
63.6590(
b)(
1)
or
§
63.6590(
b)(
2).
One
could
argue
that
a
5
HP
gasoline
generator
at
a
major
source
is
subject
to
the
initial
notification
requirements
because
it
fits
into
the
subcategory
for
emergency/
limited
use
engines.
In
the
same
manner,
it
is
not
clear
whether
an
existing
emergency
power
unit
that
is
driven
by
a
CI
engine
is
subject
to
the
requirements
of
§
63.6590(
b)(
1)
or
§
63.6590(
b)(
2)
because
the
unit
fits
into
two
categories.

Response:
The
EPA
has
clarified
the
overlap
between
RICE
subcategories
in
the
final
rule.
For
example,
an
emergency
power
unit
with
a
nameplate
rating
of
400
HP
would
not
be
14
subject
to
the
rule
since
it
is
below
500
HP.
Any
engine
rated
at
less
than
or
equal
to
500
HP
is
not
covered
by
the
final
rule
and
does
not
have
any
requirements
under
the
final
rule.
An
existing
emergency
power
unit
that
is
driven
by
a
CI
engine
would
not
be
subject
to
any
requirements
of
the
rule
since
it
is
an
existing
emergency
power
unit.
An
existing
emergency
power
unit
greater
than
500
HP
that
is
driven
by
a
4
stroke
rich
burn
(
4SRB)
engine
would
not
be
subject
to
any
requirements
of
the
rule
either.

B.
Applicability
Cutoff
I.
B.
1
Comment:
Eight
commenters
(
542,
549,
551,
554,
583,
556,
580,
587)
expressed
that
EPA
should
include
an
alternative
applicability
criteria
based
on
1
ton
per
year
(
tpy)
actual
formaldehyde
emissions.

Response:
The
basis
for
this
comment
is
the
Oil
and
Natural
Gas
Production
and
Natural
Gas
Transmission
and
Storage
NESHAP
(
promulgated
on
June
17,
1999).
In
that
rule,
hazardous
air
pollutants
(
HAP)
emissions
from
process
vents
at
glycol
dehydration
units
that
are
located
at
major
HAP
sources
and
from
process
vents
at
certain
area
source
glycol
dehydration
units
are
required
to
be
controlled
unless
the
actual
flowrate
of
natural
gas
in
the
unit
is
less
than
85,000
cubic
meters
per
day
(
3.0
million
standard
cubic
feet
per
day),
on
an
annual
average
basis,
or
the
benzene
emissions
from
the
unit
are
less
than
0.9
megagrams
per
year
(
1
tpy).
The
1
tpy
emission
threshold
in
the
Oil
and
Natural
Gas
Production
and
Natural
Gas
Transmission
and
Storage
MACT
is
equivalent
to
the
smallest
size
glycol
dehydration
unit
with
control
of
HAP
emissions
and
is,
therefore,
based
on
equivalence,
not
risk.
The
information
in
the
docket
does
not
support
a
decision
to
provide
an
alternative
applicability
cutoff
in
this
case.
The
EPA's
decision
to
defer
regulation
of
engines
500
HP
or
less
was
based
on
questions
regarding
how
accurately
the
database
reflected
such
engines.
There
was
no
such
concerns
raised
based
on
whether
an
engine
emitted
formaldehyde
above
1
tpy.

I.
B.
2
Comment:
Six
commenters
(
542,
549,
551,
554,
556,
587)
stated
that
the
applicability
limit
for
2SLB
should
be
increased
to
1,100
HP
to
be
consistent
with
the
MACT
floor.
One
commenter
(
565)
stated
that
the
small
engine
size
cutoff
should
be
changed
from
500
HP
to
650
HP.
The
commenter
said
that
while
EPA
appropriately
reasoned
that
small
engines
should
not
be
subject
to
the
requirements
of
the
rule,
EPA
provided
no
explicit
rationale
for
the
selection
of
500
HP
as
the
appropriate
small
engine
size
cutoff.
Ranking
all
engines
in
EPA's
database
from
smallest
to
largest,
the
first
engine
size
that
has
controls
is
650
HP.
Thus,
the
appropriate
small
engine
size
cutoff
supported
by
the
record
is
less
than
650
HP
instead
of
less
than
or
equal
to
500
HP.

One
commenter
(
588)
did
not
agree
that
4SLB
engines
less
than
500
HP
are
not
regarded
as
significant
sources
of
emissions.
Since
1998,
Wyoming
has
permitted
approximately
350
571
F18GL
engines
rated
at
400
HP.
Of
this
number,
218
or
63
percent
have
permit
conditions
requiring
the
installation
of
oxidation
catalyst
to
control
formaldehyde.
The
commenter
contends
that
4SLB
engines
greater
than
350
HP,
located
at
major
sources
should
be
covered
under
the
15
MACT
standard.

Response:
First,
EPA
needs
to
clarify
that
engines
500
brake
HP
or
less
have
not
been
exempted
from
regulation.
Because
EPA
determined
at
the
time
of
proposal
that
it
did
not
have
enough
information
to
go
forward
with
regulation
of
those
engines
at
this
time,
EPA
has
deferred
regulatory
activity
with
regard
to
those
engines.
Pursuant
to
a
consent
decree
signed
on
May
22,
2003,
Sierra
Club
v.
Whitman,
Case
Number
1:
01CV01537
(
D.
C.
D.
C),
a
notice
of
proposed
rulemaking
regarding
regulation
of
these
engines
under
CAA
section
112
is
scheduled
for
October
31,
2006,
with
a
final
rule
by
December
20,
2007.
At
this
time,
it
would
be
inappropriate
to
speculate
on
what
level
of
control
would
be
promulgated
for
these
engines.

The
EPA
is
aware
of
stationary
engines
as
small
as
650
HP
that
are
equipped
with
add­
on
HAP
control
devices.
The
EPA
feels
the
database
represented
the
population
of
engines
between
500
HP
and
1,100
HP
reasonably
well,
so
EPA
does
not
feel
it
is
appropriate
to
defer
regulation
of
these
engines
to
a
later
rule.
Therefore,
EPA
does
not
feel
it
is
appropriate
to
defer
the
regulation
of
engines
up
to
1,100
HP
for
2SLB
engines,
or
to
include
such
engines
in
a
separate
subcategory.
Although
650
HP
is
the
smallest
size
unit
that
is
known
to
have
add­
on
HAP
control,
EPA
feels
it
is
appropriate
to
limit
the
deferral
to
engines
500
HP
or
less
because
the
control
technology
used
for
650
HP
units
can
be
transferred
to
units
at
least
as
small
as
500
HP
in
size.
Oxidation
catalyst
technology
is
not
limited
to
engines
greater
than
650
HP
in
size.
In
fact,
information
received
during
the
public
comment
period
supports
our
conclusion,
where
several
engines
rated
at
400
HP
were
equipped
with
oxidation
catalyst
control.
The
EPA's
deferral
of
engine
regulation
was
based
on
the
type
of
engines
used
below
500
HP
and
whether
our
database
was
adequate
for
such
engines.
The
EPA
feels
the
database
for
engines
above
500
HP
was
adequate
and
that,
in
any
case,
the
final
rule
for
these
engines
are
adequately
justified
in
the
record.
The
commenter
does
not
adequately
provide
particular
reasons
to
justify
placing
engines
between
500
and
650
HP
in
a
different
subcategory
from
larger
engines,
and
EPA
does
not
feel
such
subcategorization
has
been
shown
to
be
appropriate.

The
EPA
acknowledges
the
information
provided
by
the
commenter
regarding
the
Wyoming
engines.
This
information
could
well
be
relevant
when
EPA
reviews
engines
500
HP
and
below
and
engines
at
area
sources
in
the
later
rule.

I.
B.
3
Comment:
One
commenter
(
555)
asserted
that
the
rule
should
be
more
explicit
as
to
whether
the
500
HP
capacity
level
for
exception
from
the
rule
and
5,000
HP
capacity
level
for
enhanced
monitoring
applies
to
an
individual
engine
or
applies
to
the
aggregate
capacity
of
a
group
of
engines.

Response:
The
EPA
intended
for
the
500
HP
capacity
level
to
apply
to
an
individual
engine,
not
the
aggregate
capacity
of
a
group
of
engines.
Similarly,
the
5,000
HP
capacity
level
for
enhanced
monitoring
was
intended
to
apply
to
an
individual
engine.
However,
EPA
has
not
included
a
carbon
monoxide
(
CO)
continuous
emissions
monitoring
system
(
CEMS)
requirement
in
the
final
rule.
Sources
are
free
to
use
CO
CEMS
to
demonstrate
compliance;
however,
CO
16
CEMS
are
not
required.

I.
B.
4
Comment:
One
commenter
(
527)
questioned
if
the
rule
is
reasonable
for
engines.
The
commenter
indicated
that
studies
show
that
formaldehyde
constitutes
90
percent
or
more
of
the
total
HAP
emissions.
The
commenter
stated
that
this
means
that
natural
gas
fueled
engines
total
HAP
emissions
cannot
exceed
11.1
tpy.

Response:
The
EPA
contacted
the
commenter
for
additional
information
regarding
this
comment
and
to
clarify.
The
commenter
stated
that
with
respect
to
the
rule
for
HAP,
the
commenter
noted
that
a
source
would
not
be
covered
by
the
final
rule
if
it
emitted
less
than
25
tpy
of
all
types
of
materials
considered
to
be
HAP.
This
means
that
a
site
could
emit
9
tpy
of
formaldehyde,
8
tpy
of
a
second
HAP,
and
7
tpy
of
a
third
HAP
(
totaling
24
tpy)
and
still
be
below
the
rule
limits.
Numerous
tests
of
emissions
from
natural
gas
engines
have
shown
that
these
units
emit
primarily
the
formaldehyde
HAP
and
very
little
other
HAP.
This
is
a
direct
result
of
combustion
chemistry
and
is
not
likely
to
change
by
changes
of
operating
conditions.
The
commenter
further
stated
that
nevertheless,
the
rule
regulated
sources
that
emit
a
single
species
of
HAP
at
10
tpy,
which
the
commenter
claims
places
an
additional
restriction
on
natural
gas
engines
compared
to
other
industrial
sites
that
emit
several
HAP.
The
origin
of
the
11.1
tpy
figure
can
from
using
an
estimated
emissions
mix
of
90
percent
formaldehyde
and
10
percent
other
HAP
to
estimate
the
maximum
total
HAP
that
could
be
emitted
by
a
natural
gas
engine
site
(
10
tpy
+
10
tpy
x
10
percent/
90
percent
=
11.1
tpy.
The
commenter
therefore
proposed
that
the
rule
for
the
maximum
emissions
of
a
single
HAP
be
amended
to
enable
the
maximum
emissions
by
a
single
HAP
to
be
20
tpy
for
the
cases
of
engines
using
natural
gas
fuel.
The
commenter
concluded
by
stating
that
because
of
combustion
chemistry,
such
a
rule
would
still
ensure
the
site
maximum
would
not
exceed
the
25
tpy.

The
EPA
is
required
by
section
112
of
the
CAA
to
list
categories
and
subcategories
of
major
sources
and
area
sources
of
HAP
and
to
establish
NESHAP
for
the
listed
source
categories
and
subcategories.
The
final
rule
is
limited
to
major
sources.
The
CAA
defines
major
sources
of
HAP
to
be
those
that
have
the
potential
to
emit
greater
than
10
tpy
of
any
one
HAP
or
25
tpy
of
any
combination
of
HAP.
Moreover,
the
determination
of
whether
a
source
is
a
major
source
is
made
by
counting
the
emissions
of
all
stationary
sources
located
within
a
contiguous
area
and
under
common
control.
Therefore,
EPA
cannot
look
at
the
emissions
of
an
engine
in
isolation
when
determining
whether
it
is
located
at
a
major
source.
The
suggested
revision
and
the
logic
underlying
the
comment
are
therefore
not
consistent
with
the
specific
requirements
of
the
CAA.
I.
B.
5
Comment:
One
commenter
(
565)
stated
that
EPA
retains
the
authority
to
establish
applicability
criteria
that
focus
the
rule
on
the
emission
units
emitting
HAP
using
de
minimis
criteria.
Despite
the
National
Lime
II
and
CKRC
cases,
EPA
has
not
lost
any
of
its
discretion
to
establish
applicability
criteria
that
define
the
source
category
and
the
units
subject
to
a
MACT
standard.
The
EPA
should
revisit
its
decision
that
it
does
not
have
de
minimis
authority.
The
EPA's
interpretation,
discussed
in
National
Lime
II,
that
it
lost
its
inherent
authority
to
relieve
burdens
of
regulation
where
they
yield
trivial
or
no
value
is
based
on
an
incorrect
analysis
of
the
wrong
provision
of
the
statute.
The
commenter
suggested
that
EPA
should
not
take
such
a
17
restrictive
position
on
its
de
minimis
authority.
The
commenter
suggested
that
EPA
reexamine
its
policy
that
de
minimis
exclusions
are
unavailable
when
EPA
calculates
a
floor,
see
National
Lime
II,
supra
at
640,
because
that
policy
was
based
on
an
incorrect
analysis
of
the
statute.
Even
if
EPA
does
not
alter
that
policy,
the
commenter
noted
that
it
does
not
affect
EPA's
inherent
authority
to
craft
de
minimis
exclusions
unrelated
to
setting
the
MACT
floor.

Response:
The
EPA
thanks
the
commenter
for
its
analysis.
However,
as
the
commenter
does
not
relate
the
analysis
to
any
particular
decision
of
the
Agency
in
the
proposed
rule,
the
EPA
does
not
feel
that
it
is
necessary
or
appropriate
to
respond
to
the
comment
here.
It
would
be
more
appropriate
to
review
the
legal
analysis
where
it
can
be
applied
to
a
particular
set
of
facts
that
are
relevant
to
the
particular
regulatory
action
being
addressed.

C.
Exemptions
I.
C.
1
Comment:
Two
commenters
(
526,
579)
requested
that
EPA
exempt
units
(
new
and
existing)
with
non­
selective
catalytic
reduction
(
NSCR)
or
oxidation
catalyst
that
have
been
installed
in
compliance
with
other
regulatory
mandates
to
avoid
potential
conflict.

Response:
The
EPA
does
not
agree
with
the
commenters
who
suggested
an
exemption
for
existing
and
new
units
with
NSCR
or
oxidation
catalyst
controls
that
have
been
installed
in
compliance
with
other
regulations.
The
EPA
has
therefore
not
exempted
units
with
NSCR
or
oxidation
catalyst
controls
that
have
been
installed
in
compliance
with
other
regulations
from
the
final
rule.
These
units
are
covered
and
must
demonstrate
compliance.
The
EPA
may
look
at
the
rule
again
at
a
later
date
following
promulgation
if
some
subgroup
of
new
engines
do
not
appear
to
be
able
to
meet
the
standards
either
through
the
designated
control
technique
or
through
other
types
of
reasonable
controls,
but
such
evidence
does
not
exist
in
this
record.
For
existing
engines,
the
Act
understood
that
some
engines
are
already
controlled
may
not
meet
the
requirements
of
the
MACT,
which
is
designed
to
achieve
the
control
achieved
by
the
average
of
the
best
controlled
12
percent
of
engines.

I.
C.
2
Comment:
One
commenter
(
604)
contended
that
the
MACT
should
consider
exempting
any
RICE
using
landfill
gas.
A
diesel
engine
can
operate
at
a
landfill
in
a
dual­
fuel
mode
using
fuel
oil
and
landfill
gas.
Tests
have
shown
that
a
catalytic
converter
cannot
be
used
because
of
siloxanes
in
the
landfill
gas,
even
if
the
engine
operates
with
more
than
half
the
energy
being
supplied
by
the
liquid
fuel.

Six
commenters
(
526,
544,
547,
565,
579,
584)
expressed
support
for
EPA's
approach
to
deal
with
engines
with
landfill
gas.
Commenters
526,
579,
and
584
agreed
that
fuel
switching
is
an
environmentally
inferior
option.
Commenter
584
agreed
with
EPA
that
emissions
reductions
for
landfill
gas
powered
engines
are
not
required.

Response:
In
the
proposed
rule,
EPA
had
established
a
subcategory
for
landfill
and
digester
gas­
fired
units
and
defined
the
subcategory
as
those
stationary
RICE
that
combust
18
digester
gas
or
landfill
gas
as
the
primary
fuel.
In
the
proposed
rule,
these
units
did
not
have
to
meet
any
emission
limitation
requirements
but
were
subject
to
the
initial
notification
requirements.
The
EPA
agrees
with
the
commenters
supporting
the
proposed
approach
to
landfill
and
digester
gas
fired
engines.
The
EPA
agrees
that
neither
control
technology,
fuel
switching,
or
other
practices
would
be
an
appropriate
or
workable
strategy
for
reducing
HAP
from
these
engines.
The
EPA
agrees
with
the
commenter
that
problems
will
occur
when
using
landfill
gas
because
of
siloxanes
in
the
fuel,
even
if
the
engine
operates
with
more
than
half
the
energy
being
supplied
by
the
liquid
fuel.
Therefore,
EPA
contacted
sanitation
districts
and
catalyst
vendors
for
information.
Based
on
the
information
obtained,
EPA
feels
that
firing
greater
than
10
percent
landfill
or
digester
gas
will
cause
fouling
of
the
oxidation
catalyst,
rendering
the
control
device
inoperable
within
a
short
period
of
time.
All
the
sources
EPA
contacted
indicated
that
there
would
be
problems
associated
with
catalyst
deactivation
due
to
siloxanes
present
in
landfill
and
digester
gas.
For
further
information
please
refer
to
the
memorandum
entitled
"
Stationary
Combustion
Turbines
and
Reciprocating
Internal
Combustion
Engines
fired
with
Landfill
or
Digester
Gas"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35)).
While
most
units
will
operate
using
landfill
or
digester
gas
consumption
above
50
percent
of
the
time,
there
are
times
when
such
units
may
need
to
operate
significantly
below
50
percent
landfill
or
digester
gas
consumption.
The
EPA
feels
a
cutoff
level
of
10
percent
of
gross
heat
input
is
an
appropriate
level
for
defining
these
units,
because
operation
below
that
percentage
raises
significant
questions
regarding
whether
the
unit
is
still
appropriately
considered
operating
as
a
landfill
or
digester
gas
burning
unit,
and
would
raise
concerns
regarding
circumvention
of
the
requirements
for
other
new
units.
In
the
final
rule,
EPA
has
defined
the
subcategory
as
those
engines
with
annual
landfill
and
digester
gas
consumption
of
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis.
Engines
in
this
subcategory
must
only
comply
with
limited
requirements
of
the
final
rule.
Engines
with
an
annual
landfill
and
digester
gas
consumption
of
less
than
10
percent
of
the
gross
heat
input
on
an
annual
basis
are
subject
to
applicable
emission
limitations
of
the
final
rule
in
addition
to
other
requirements.

I.
C.
3
Comment:
One
commenter
(
605)
requested
that
EPA
consider
the
use
of
pre­
gas
treatment
for
digester
and
landfill
gas­
fired
engines,
thereby
allowing
use
of
oxidation
catalysts
for
HAP
reduction.

Response:
The
EPA
investigated
the
use
of
pre­
gas
treatment
for
digester
and
landfill
gas­
fired
engines.
The
EPA
considered
a
pretreatment
system.
However
the
cost
of
such
systems
were
deemed
prohibitive.
There
is
also
very
limited
experience
in
the
use
of
pretreatment
systems
to
remove
siloxanes.
As
a
result,
there
is
insufficient
information
on
the
siloxane
removal
effectiveness
of
pretreatment
systems.
The
EPA
is
aware
of
pretreatment
systems
being
applied
to
certain
facilities.
However,
more
testing
is
required
to
verify
the
long­
term
performance
of
the
pretreatment
system
and
its
effect
on
catalysts.
Please
refer
to
the
memorandum
entitled
"
Control
Technology
Options
for
Existing
Digester
Gas
and
Landfill
Gas
Engines,"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35))
and
the
memorandum
entitled
"
Stationary
Combustion
Turbines
and
Reciprocating
Internal
Combustion
Engines
fired
with
Landfill
or
Digester
Gas"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
19
0059
(
A­
95­
35))
for
more
information.

I.
C.
4
Comment:
Two
commenters
(
549,
551)
asked
for
clarification
on
whether
the
rule
intends
to
cover
stationary
internal
combustion
engines
at
refineries
that
are
used
for
fuel
testing.
It
is
the
commenters'
understanding
that
this
rule,
as
opposed
to
the
Engine
Test
Cell/
Stand
MACT
governs
those
sources,
as
applicable.

Response:
Stationary
internal
combustion
engines
at
refineries
that
are
used
for
fuel
testing
are
covered
by
the
final
rule.
Knock
engines
are
used
at
refineries,
and
do
not
test
the
engine
per
se
but
instead
test
the
fuels
and
lubricants
for
product
quality
and
development
purposes.
They
are
therefore
covered
by
this
rule
as
opposed
to
the
Engine
Test
Cell/
Stand
MACT.
The
engines
are
typically
smaller
units
and
are
similar
to
automotive
engines
commonly
ranging
from
100
to
200
HP
in
size.
If
the
units
used
for
fuel
testing
are
500
HP
or
less
they
would
not
be
covered
by
the
final
rule.

I.
C.
5
Comment:
Four
commenters
(
544,
547,
565,
569)
expressed
support
for
the
Agency's
decision
not
to
require
emissions
limitations
for
existing
2SLB,
existing
4SLB,
existing
CI,
emergency
power/
limited
use,
and
500
HP
or
less.
One
commenter
(
549)
supported
EPA
setting
a
"
No
Control"
floor
and
exemption
for
existing
2SLB,
4SLB,
and
CI
stationary
RICE.
One
commenter
(
565)
also
supported
the
exemption
for
stationary
RICE
test
cells
and
stands.
Commenter
575
agreed
with
the
decision
to
subcategorize
very
small
sources
(
500
HP
or
less)
and
to
defer
the
standard­
setting
decision,
in
light
of
the
scarcity
of
data
and
apparently
trivial
emissions
attributable
to
these
units.
One
commenter
(
553)
expressed
support
for
EPA's
exclusion
of
emergency
and
limited
use
engines.

One
commenter
(
575)
endorsed
EPA's
proposal
to
establish
an
emergency
power
subcategory
for
stationary
RICE,
with
emergency
defined
with
reference
to
the
use
of
a
stationary
RICE
as
a
power
source
when
the
primary
power
source
for
a
facility
has
been
rendered
inoperable
by
an
emergency
situation
(
without
reference
to
a
specific
number
of
hours).
The
commenter
also
stated
that
the
decision
that
the
MACT
floor
is
no
emissions
reduction
is
appropriate.

Response:
The
EPA
acknowledges
the
commenters'
support.
Also
note
that
engines
less
than
or
equal
to
500
HP
are
not
covered
by
the
final
rule.
The
EPA
is
covering
area
and
small
sources
in
a
later
rule.

I.
C.
6
Comment:
One
commenter
(
594)
stated
that
the
proposal
that
the
MACT
floor
for
existing
CI
be
no
emissions
reductions
is
endorsed.
The
finding
that
less
than
1
percent
of
existing
CI
have
controls
is
consistent
with
the
commenter's
experience
and
supports
the
floor
finding.

Response:
The
EPA
has
noted
the
commenter's
support.

I.
C.
7
Comment:
One
commenter
(
584)
stated
that
EPA
should
not
develop
NESHAP
20
for
engines
that
utilize
landfill
gas
because
these
engines
are
one
of
the
control
technologies
mandated
by
the
municipal
solid
waste
landfill
NESHAP.
As
such,
EPA
should
not
be
regulating
control
devices
with
control
devices.

Response:
The
commenter
provided
no
legal
support
for
its
position
that
a
pollution
source
should
be
exempt
from
CAA
section
112(
d)
because
it
may
have
been
installed
to
comply
with
other
requirements.
In
any
case,
the
commenter
otherwise
agreed
with
EPA's
approach
to
regulating
engines
operating
on
landfill
or
digester
gas,
which
are
not
required
to
meet
emission
limitations.

D.
Emergency/
Limited
Use
I.
D.
1
Comment:
One
commenter
(
544)
stated
that
the
limited
use
definition
for
RICE
should
be
separated
from
the
emergency
power
definition
since
these
are
really
different
applications.
Another
commenter
(
565)
said
that
including
both
types
of
units
in
the
same
definition
creates
challenges
in
determining
which
criteria
apply
to
each
use
and
how
to
characterize
a
unit
that
may
have
both
functions.
Commenter
565
suggested
that
for
clarity,
the
descriptions
of
each
type
of
unit
may
require
separation
in
the
rule.
Second,
in
defining
each
unit
type
EPA
needs
to
also
address
related
functions
performed
on
each
unit
type
that
are
required
to
support
the
primary
function
of
the
unit.
This
commenter
also
stated
that
EPA
should
clarify
the
application
of
the
emergency
power/
limited
use
definition
to
multi­
use
units
(
both
emergency
use
and
limited
non­
emergency
use).

Twenty­
three
commenters
(
536,
540,
542,
544,
545,
547,
548,
551,
552,
553,
554,
556,
560,
562,
565,
566,
571,
573,
587,
593,
594,
595,
603(
D))
stated
that
a
limited
use
category
should
be
included.
One
commenter
(
536)
suggested
using
a
flat
annual
threshold
level
of
1,000
hrs/
yr
in
lieu
of
10
percent
usage.
Another
commenter
(
545)
recommended
that
the
category
include
all
units,
not
only
peak
shaving
units.
One
commenter
(
551)
recommended
that
this
category
should
include
all
periods
of
testing
and
maintenance.
Two
commenters
(
548,
560)
stated
that
in
the
event
that
a
limited
use
RICE
exceeds
876
hours
in
1
year,
EPA
should
allow
1
year
for
the
owner
or
operator
to
meet
the
MACT
standard.
One
commenter
(
595)
agreed
with
EPA
that
limited
use
RICE
are
appropriate
for
subcategorization
because
"
their
use
and
operation
are
different
compared
to
typical
RICE."
Two
commenters
(
544,
565)
suggested
a
10
percent
annual
capacity
factor
for
limited
use.

Twenty­
seven
commenters
(
526,
540,
542,
544,
547,
548,
549,
552,
554,
555,
556,
557,
558,
559,
560,
564,
565,
569,
571,
573,
575,
576,
577,
579,
581,
858,
587,
595)
argued
that
the
50
hrs/
yr
may
not
be
sufficient.
Six
commenters
(
540,
558,
569,
576,
577,
581)
noted
that
testing
and
maintenance
should
be
included
and
not
counted
towards
the
50
hrs/
yr.
Two
commenters
(
548,
560)
recommended
at
least
250
hrs/
yr.
In
the
event
that
an
engine
exceeds
250
hours
in
1
year,
EPA
should
allow
1
year
for
the
owners
and
operators
to
meet
the
standard.
One
commenter
(
549)
stated
that
the
number
of
hours
for
emergency/
limited
use
exclusion
should
be
increased
to
provide
for
periodic
reliability
and
maintenance
testing.
One
commenter
(
560)
21
recommended
a
52
hour
limit
for
routine
maintenance
and
testing,
then
have
no
limit
for
true
emergency
use.
Similarly,
commenters
544
and
565
expressed
that
since
routine
or
unscheduled
maintenance
and
testing
could
require
unknown
time
to
complete,
there
should
be
no
time
limits
on
the
use
of
emergency
stationary
RICE.
Another
commenter
(
555)
contended
that
52
hrs/
yr
is
satisfactory
for
most
engines.
However,
engines
located
at
airports
can
require
up
to
100
hrs/
yr
and
engines
at
nuclear
power
plants
can
require
up
to
200
hrs/
yr.
Seven
commenters
(
542,
552,
554,
556,
571,
573,
587)
suggested
100
hrs/
yr
for
emergency
generators.
Two
commenters
(
569,
564)
recommended
that
the
50
hours
should
be
extended
to
10
percent.
One
commenter
(
547)
asked
that
EPA
clarify
that
testing
is
allowed
as
an
emergency
use.
Two
commenters
(
549,
559)
recommended
that
the
50
hrs/
yr
definition
be
changed
to
mean
an
engine
whose
operation
is
restricted
by
federally
enforceable
limits
to
less
than
876
hrs/
yr.
One
commenter
(
595)
stated
that
the
subcategory
should
be
redefined
to
include
RICE
that
operate
less
than
500
hrs/
yr.
In
the
event
that
an
emergency
power
RICE
exceeds
500
hours
in
1
year,
EPA
should
allow
1
year
for
the
owners
and
operators
to
meet
the
MACT
standard,
if
necessary.
The
500
hours
of
operation
would
be
consistent
with
Ohio
Administrative
Code
3745­
31­
03
(
A)(
4)
which
grants
emergency
generators
greater
than
50
HP
a
permit­
by­
rule
permit
to
install
exemption
if
the
unit
operates
less
than
500
hours
per
rolling
12
month
period.
Two
commenters
(
526,
579)
remarked
that
setting
this
exemption
at
50
hrs/
yr
down
from
the
100
or
200
hrs/
yr
commonly
seen
in
many
State
air
pollution
regulations,
could
have
the
net
effect
of
increasing
pollution
by
not
allowing
sufficient
operating
time
for
the
engine
to
burn
off
hard
deposits.
One
commenter
(
585)
expressed
that
testing
of
engines
be
excluded
from
the
50
hrs/
yr
or
the
threshold
limit
of
this
subcategory
be
changed
from
50
hrs/
yr
to
be
10
percent
of
the
annual
capacity
factor.
One
commenter
(
575)
asked
that
these
units
should
be
allowed
to
run
for
normal
maintenance
purposes.
Commenter
573
suggested
that
the
testing
of
engines
be
excluded
from
the
definition
or
the
threshold
limit
of
the
subcategory
be
changed
from
50
hours
to
be
10
percent
of
the
annual
capacity
factor.
One
commenter
(
557)
urged
further
study
of
the
stringent
testing
time
guidelines
suggested
and
refrain
from
implementation
until
such
data
can
be
reviewed
and
offered
in
a
proper
consensus
forum.

Two
commenters
(
544,
565)
stated
that
the
operation
of
emergency
power
units
should
not
be
limited
to
only
those
times
when
the
primary
power
source
has
been
interrupted,
but
rather
not
time­
restricted
at
all,
providing
the
primary
design
purpose
of
the
unit
is
to
provide
emergency
backup
services,
fire
water,
etc.

One
commenter
(
555)
asked
that
EPA
clarify
the
definition
of
emergency/
limited
use
engines
as
to
whether
loss
power
that
constitutes
an
emergency
is
limited
to
power
supplied
to
the
facility
as
a
whole
or
includes
power
supplied
to
portions
of
the
facility.

One
commenter
(
552)
suggested
that
EPA
revise
the
definition
of
emergency
power
RICE
to
clarify
the
intent
of
the
rule
as
the
current
definition
does
not
adequately
encompass
the
wide
array
of
emergency
uses
of
engines.

One
commenter
(
594)
endorsed
the
proposed
subcategory
of
no
emission
reduction
for
RICE
associated
with
emergency
power/
limited
use
and
the
specific
allowance
for
50
hours
of
incidental
22
use
in
non­
emergency
mode.

One
commenter
(
565)
felt
that
the
description
of
an
emergency
engine
is
too
restrictive.
The
emergency
use
description
should
describe
more
power
loss
emergencies
than
those
affecting
an
entire
facility
at
once.
The
commenter
stated
that
EPA
should
clarify
that
emergency
use
also
occurs
when
a
standby
or
emergency
unit
activates
to
replace
the
function
or
perform
an
emergency
function
that
arises
from
the
sudden
inoperability
of
one
or
more
pieces
of
equipment
on
a
site.
The
definition
should
also
include
uses
for
additional
emergency
types
beyond
power
loss
emergencies,
e.
g.,
fuel
and
raw
material
curtailments
or
fuel
shortage
emergencies
applied
by
governments,
utilities,
or
other
suppliers
may
require
the
need
to
temporarily
operate
an
engine,
or
some
equipment
may
be
operated
to
fight
fires
(
firewater
pumps).
Neither
of
these
examples
represent
loss
of
power,
but
are
still
unplanned
events.

One
commenter
(
577)
requested
that
EPA
provide
further
clarification
on
the
term
to
include
provisions
for
scenarios
such
as
power
fluctuations
and
requests
by
local
utility
to
run
their
generators.

One
commenter
(
576)
said
that
when
healthcare
facilities
operate
their
emergency
generators
due
to
severe
weather
conditions
or
inability
of
the
local
utility
to
provide
acceptable
power,
healthcare
organizations
should
be
allowed
to
operate
their
generators
in
support
of
equipment/
systems
critical
to
patient/
resident
care.

One
commenter
(
573)
noted
that
the
definition
should
be
revised
to
include
operation
at
control
centers
for
human
spaceflight
to
provide
power
in
and
to
allow
for
operation
due
to
foreseeable
but
critical
electrical
outages
or
events
anticipation
of
power
failure
due
to
severe
storm
activity.

One
commenter
(
555)
stated
that
the
definition
should
be
clarified,
or
extended,
to
allow
for
operations
in
anticipation
of
an
emergency
situation.

One
commenter
(
567)
remarked
that
this
class
of
RICE
(
engines
having
a
capacity
utilization
of
less
than
10
percent)
would
operate
mostly
in
the
summer
months
when
the
public
is
more
likely
to
be
impacted
by
the
emissions.
Acetaldehyde,
acrolein,
and
formaldehyde
all
have
documented
short­
term
acute
health
effects.
The
EPA
has
failed
to
identify
short­
term
health
effects
throughout
any
of
the
risk
analysis
proposals.
The
commenter
asserted
that
any
subcategorization
of
these
engines
without
controls
is
not
protective
of
public
health.

One
commenter
(
566)
contended
that
the
subcategory
of
limited
use
must
be
tightly
defined
in
terms
of
capacity
and
utilization,
rather
than
the
descriptive
term
"
peak
shaving,"
since
limited
use
stationary
RICE
will
perform
functions
other
than
just
peak
shaving.
Since
EPA
has
determined
that
these
units
will
likely
have
a
floor
of
no
emissions
reductions,
it
would
not
be
appropriate
to
require
anything
other
than
an
initial
notification
and
a
requirement
to
submit
annual
records
of
total
operating
hours.
23
One
commenter
(
567)
said
that
any
subcategorization
of
RICE
operated
less
than
876
hrs/
yr
should
include
a
permit
requirement
that
these
units
operate
less
than
876
hrs/
yr.
The
commenter
further
noted
that
this
class
could
have
less
onerous
monitoring
requirements
as
a
way
to
lower
costs
for
these
engines
especially
when
the
low
amount
of
operating
hours
are
considered.

One
commenter
(
553)
expressed
support
for
an
"
emergency
use"
subcategory
that
adds
to
the
definition
of
"
emergency
use"
units
as
alternative
to
limited
use
category.
The
commenter
strongly
supported
an
"
emergency
use"
subcategory
that
construes
"
emergency
use"
more
broadly
to
include
the
use
of
peak­
shaving
units
during
periods
of
high
demand.
Inevitably,
during
periods
of
peak
demand
when
nearly
all
the
available
generating
units
are
dispatched,
"
capacity
alerts"
are
declared.
The
alerts
are
issued
with
varying
degrees
of
severity.
The
commenter
supported
the
inclusion
of
"
generation
emergencies"
in
its
definition
of
"
emergency
use."
Specifically,
the
commenter
supported
a
determination
by
EPA
that,
at
a
particular
level
of
severity,
the
alerts
constitute
an
"
emergency,"
for
which
there
is
no
limitation
in
terms
of
hours
of
percentage
of
capacity
used.

One
commenter
(
544)
argued
that
the
limited
use
definition
for
RICE
should
be
separated
from
the
emergency
power
definition
since
these
are
really
different
applications.

One
commenter
(
585)
stated
that
the
definition
of
emergency
should
be
expanded
to
include
nonemergency
hours
of
operation
that
are
supportive
of
emergency
situations.
In
addition,
the
definition
should
not
have
any
time
limits
specified
on
emergency
RICE
units.

One
commenter
(
544)
indicated
that
the
definition
needs
to
allow
for
periodic
testing
of
the
RICE
unit
to
ensure
that
it
is
reliable
and
available
should
an
emergency
occur,
in
addition
to
the
time
it
may
be
required
to
provide
emergency
service.

One
commenter
(
540)
supported
that
the
definition
of
emergency
engine
does
not
limit
the
number
of
emergency
hours
allowed
for
the
unit.

One
commenter
(
593)
stated
that
the
definition
of
emergency
power/
limited
use
needs
to
be
changed
as
it
does
not
allow
enough
operating
time
for
non­
emergency
use
of
RICE
and
does
not
cover
all
the
emergency
uses
of
RICE.
The
definition
should
include
non­
emergency
hours
of
operation
that
are
supportive
of
emergency
mode
operations.
For
example,
the
emergency
use
should
also
cover
the
periodic
testing
of
the
RICE
unit
to
ensure
that
it
is
reliable
and
available
should
an
emergency
occur.
The
current
language
appears
to
limit
these
test
periods
to
less
than
50
hours
per
year.
The
EPA
should
clarify
that
emergency
use
occurs
when
a
standby
or
emergency
unit
activates
to
replace
the
function
or
perform
an
emergency
function
that
arises
from
the
sudden
inoperability
of
one
or
more
pieces
of
equipment
on
a
site.
The
commenter
recommended
modifying
the
definition
to
include
periods
of
operation
in
response
to
plant
emergency
situations
whether
or
not
primary
power
is
lost.

One
commenter
(
564)
suggested
eliminating
from
the
definition
the
reference
to
"
when
the
24
primary
power
source
has
been
rendered
inoperable."
There
are
emergency
conditions
where
the
primary
power
source
is
still
operable,
but
the
emergency
condition
necessitates
the
startup
of
engines
(
e.
g.,
firewater
pumps
during
a
unit
fire,
instrument
air
back­
up
engines).
Another
option
would
be
to
add
the
words
"
or
is
insufficient
for
an
emergency
situation"
after
the
primary
power
source
comment.
In
addition,
the
commenter
stated
that
the
definition
for
emergency
RICE
should
include
the
testing
of
emergency
engines
to
ensure
their
availability
and
reliability
for
emergency
situations.

One
commenter
(
548)
remarked
that
the
definition
of
emergency
use/
limited
power
could
be
improved
as
follows:
"
An
emergency
power/
limited
use
unit
means
any
stationary
RICE
that
operates
as
a
mechanical
or
electric
power
source
during
emergencies,
when
the
primary
power
source
for
a
facility
has
been
rendered
inoperable
by
an
emergency
situation
and
units
that
operate
less
than
250
hours
per
year
in
non­
emergency
situations
or
for
equipment
testing
and
maintenance,
including
certain
peaking
units
at
electric
facilities
or
stationary
RICE
at
industrial
facilities.
One
example
of
an
emergency
situation
is
when
electric
power
from
the
local
utility
is
interrupted.
Another
example
is
to
pump
water
in
the
case
of
fire
or
flood."

Response:
The
preamble
to
the
proposed
rule
proposed
a
subcategory
for
limited
use
stationary
RICE
and
defined
them
as
operating
50
hours
or
less
per
calendar
year.
Comments
received
indicated
that
the
proposed
50
hours
per
year
for
limited
use
units
was
not
sufficient
and
that
many
limited
use
engines
would
exceed
the
50
hours
per
year
just
by
routine
testing
and
maintenance
of
the
engine
for
readiness
purposes.
For
this
reason,
EPA
feels
that
few
owners
and
operators
would
find
this
allowance
useful
and
would
not
serve
a
purpose
except
to
cover
periods
of
testing
and
maintenance.
The
EPA
has,
therefore,
found
it
appropriate
to
increase
the
number
of
hours
for
limited
use
operation.
The
EPA
has
specified
in
the
final
rule
that
limited
use
stationary
RICE
are
stationary
RICE
that
operate
less
than
100
hours
per
year.
For
limited
use
units,
operation
during
routine
testing
and
maintenance
is
counted
towards
the
100
hours
per
year.

In
the
preamble
to
the
proposed
rule,
EPA
solicited
comments
on
creating
a
subcategory
of
limited
use
engines
with
capacity
utilization
of
10
percent
or
less
(
876
or
fewer
hours
of
annual
operation).
These
units
would
have
included
engines
used
for
electric
power
peak
shaving.
As
a
result
of
soliciting
comments,
EPA
received
several
comments
regarding
the
possibility
of
establishing
a
limited
use
subcategory
with
capacity
utilization
of
10
percent
or
less;
some
for
and
some
against.
The
EPA
considered
all
comments
received
and
has
decided
not
to
include
a
subcategory
of
limited
use
stationary
RICE
with
a
capacity
utilization
of
10
percent
or
less
in
the
final
rule.
Limited
use
units
operating
876
hours
per
year
are
similar
to
other
sources
equipped
with
add­
on
oxidation
catalyst
control
and
their
operation
only
during
peak
periods
does
not
preclude
them
from
being
equipped
with
add­
on
oxidation
catalyst
control.
Those
commenters
supporting
a
longer
time
period
for
the
limited
use
engines
did
not
provide
persuasive
arguments
for
such
a
subcategory.
The
commenters
have
not
provided
significant
data
indicating
that
engines
operating
up
to
10
percent
of
the
time
(
or
longer,
as
some
commenters
suggested)
are
unable
to
take
steps
similar
to
other
RICE
to
reduce
HAP.
On
the
contrary,
as
stated
previously,
25
such
engines
are
similar
to
other
stationary
RICE
that
can
be
and
have
been
equipped
with
add­
on
oxidation
catalyst
control,
and
their
operation
only
during
peak
periods
does
not
preclude
them
from
being
equipped
with
workable
add­
on
control
or
from
using
other
methods
of
emission
control
to
reduce
HAP.
The
10
percent
time
limit
would
allow
over
a
month
of
usage
per
year,
which
EPA
feels
is
substantial
enough
both
to
be
of
concern
environmentally
and
to
take
advantage
of
emission
control
strategies.
Significant
operation
of
these
engines
is
expected
and
should
be
accounted
for
in
the
final
rule.

By
contrast,
a
limited
use
exemption
covering
only
100
hours
per
year
of
use
is
justified
because
usage
in
these
cases
in
clearly
exceptional
and
these
engines
would
have
the
technical
and
usage
concerns
similar
to
emergency
engines
discussed
in
the
proposed
rule.
These
engines
are
categorically
different
from
other
engines
in
that
they
are
only
used
in
truly
exceptional
situations.
For
these
reason,
EPA
has
not
established
a
limited
use
subcategory
of
units
operating
876
hours
per
year
in
the
final
rule,
but
has
included
a
limited
use
subcategory
allowing
engines
to
operate
up
to
100
hours
per
year.

The
EPA
does
not
agree
with
the
commenters
who
suggested
that
in
the
event
that
a
limited
use
engine
exceeds
the
set
amount,
EPA
should
allow
1
year
for
the
owners
and
operators
to
meet
the
standard.
The
EPA
feels
that
when
limited
use
engines
are
put
into
more
than
limited
service,
these
engines
should
be
required
to
meet
applicable
standards
in
the
timeframe
specified.
Owners
and
operators
of
limited
use
engines
should
be
well
aware
of
the
possibility
of
these
engines
needing
to
be
put
into
regular
service
and
should
plan
accordingly.
Therefore,
a
1
year
compliance
timeframe
is
not
needed.

The
EPA
agrees
with
the
comment
that
the
emergency
and
limited
use
stationary
RICE
definition
should
be
separated.
The
EPA
has
established
separate
definitions
for
emergency
stationary
RICE
and
limited
use
stationary
RICE
in
the
final
rule.

In
addition,
in
the
final
rule,
the
definition
of
emergency
engine
was
written
to
indicate
that
loss
of
power
that
constitutes
an
emergency
can
include
power
supplied
to
portions
of
a
facility.
The
EPA
intended
that
the
definition
of
emergency
engine
include
operation
during
emergency
situations,
including
times
when
the
primary
power
source
has
been
interrupted
as
well
as
other
situations
such
as
pumping
water
in
the
case
of
fire
or
flood,
which
was
given
as
an
example
of
emergency
operation
in
the
definition
in
the
proposed
rule.
The
definition
has
been
clarified
to
clearly
indicate
that
emergency
operation
is
not
limited
to
only
times
when
the
primary
power
source
has
been
interrupted.
The
EPA
contacted
the
commenter
for
more
information
about
the
types
of
curtailments
with
which
they
were
concerned.
The
commenter
provided
only
one
example,
which
was
shutdown
of
offshore
wells
during
a
hurricane.
The
EPA
feels
that
the
definition
of
emergency
stationary
combustion
engine
is
sufficient
to
cover
this
particular
scenario
and
it
is
not
necessary
to
include
more
examples
of
emergency
operation.
It
would
be
nearly
impossible
to
provide
examples
of
every
potential
type
of
emergency
situation.
The
operation
of
emergency
engines
is
not
limited
to
a
specific
number
of
hours.
Also,
routine
testing
and
maintenance
to
ensure
operational
readiness
have
been
included
in
the
definition
of
emergency
26
engine.
However,
the
routine
testing
and
maintenance
must
be
within
limits
recommended
by
the
engine
manufacturer
or
other
entity
such
as
an
insurance
company.
Emergency
stationary
RICE
may
also
operate
an
additional
50
hours
per
year
in
non­
emergency
situations.
As
stated
previously,
routine
testing
and
maintenance
have
been
included
in
the
definition
of
emergency
stationary
RICE
and,
therefore,
are
not
counted
towards
the
50
hours
per
year.
The
EPA
does
not
agree
that
operation
in
anticipation
of
an
emergency
situation
should
be
included
in
the
definition
of
emergency
engine
and
has
not
included
such
an
allowance
in
the
final
rule.

In
response
to
the
commenter
who
urged
further
study
of
the
stringent
testing
time
guidelines
suggested
and
refrain
from
implementation
until
such
data
can
be
reviewed
and
offered
in
a
proper
consensus
forum,
the
EPA
believes
that
the
above
response,
which
indicates
that
maintenance
and
testing
of
emergency
engines
have
been
included
in
the
definition
of
an
emergency
engine,
satisfies
the
commenter's
concern.
There
is
no
limit
on
the
number
of
hours
allowed
for
testing
and
maintenance,
but
the
routine
testing
and
maintenance
must
be
within
limits
recommended
by
the
engine
manufacturer
or
other
entity
such
as
an
insurance
company.

I.
D.
2
Comment:
One
commenter
(
566)
requested
clarification
on
the
following
situation.
The
commenter
has
an
emergency
system
that
provides
controlled
cooling
water
to
cool
its
number
7
blast
furnace
in
case
of
a
loss
of
electrical
power
or
a
mechanical
pump
failure.
The
system
consists
of
an
elevated
water
storage
tank/
tower
and
two
600
HP
diesel
engines
that
must
be
started
before
the
storage
capacity
is
depleted
during
an
emergency
situation
to
maintain
the
minimum
required
cooling
water
flow
by
pumping
to
the
tower/
tank.
The
rule
clarification
is
required
for
the
fact
that
this
system
is
also
utilized
for
furnace
shutdowns
to
perform
required
maintenance
work
to
avoid
catastrophic
failures/
emergencies.
The
commenter
said
the
intent
of
this
operation
fits
within
the
definition
of
emergency
power/
limited
use,
but
this
is
not
clear
by
the
specific
definition.

Response:
This
specific
example
would
fall
under
the
definition
of
emergency
operation.

I.
D.
3
Comment:
One
commenter
(
547)
stated
that
some
limited
use
units
do
not
have
stacks
that
meet
the
performance
specification
requirements,
and
consequently,
emissions
testing
on
such
units
may
require
the
prohibitively
expensive
construction
of
such
stacks.
As
a
result
of
these
differences,
the
commenter
stated
that
EPA
should
create
a
separate
subcategory
for
limited
use
RICE.
The
commenter
supported
a
subcategory
of
limited
use
RICE
with
annual
capacity
utilization
of
10
percent
or
less.

Response:
The
EPA
has
included
a
provision
for
limited
use
units
in
the
final
rule
which
allows
limited
use
units
to
operate
up
to
100
hours
per
year
without
being
subject
to
the
emission
and
operating
limitation
requirements
of
the
final
rule.
The
rationale
for
defining
limited
use
as
those
units
that
operate
less
than
100
hours
per
year
as
opposed
to
those
that
operate
less
than
876
hours
per
year
was
discussed
in
the
response
to
comment
I.
D.
1.
If
the
units
the
commenter
is
concerned
with
qualify
as
limited
use
units,
which
they
may
since
EPA
has
increased
the
limited
use
amount
from
50
to
100
hours
per
year,
owners
and
operators
would
not
have
to
comply
with
27
the
emission
and
operating
limitations
and
would
not
have
to
conduct
performance
tests.
However,
it
is
not
clear
from
the
comment
whether
the
commenter
is
referring
to
performance
specifications
for
CO
CEMS
or
if
the
commenter
is
referring
to
performance
test
requirements.
The
EPA
proposed
requiring
CO
CEMS
for
large
lean
burn
and
large
CI
engines
complying
with
the
requirement
to
reduce
CO
emissions.
However,
based
on
comments
received,
EPA
is
no
longer
requiring
CO
CEMS.
Sources
have
the
option
of
using
CO
CEMS,
but
it
is
not
required.
Therefore,
CO
CEMS
performance
specifications
should
not
be
an
issue
for
these
sources.
If
these
units
are
not
limited
use
units,
EPA
cannot
exempt
them
from
performance
testing
requirements.
Section
63.7(
d)
of
the
General
Provisions
requires
owners
and
operators
to
provide
adequate
performance
testing
facilities,
including
a
stack.
The
commenter
did
not
provide
any
information
supporting
why
the
construction
of
such
stacks
would
be
prohibitively
expensive.
The
EPA
therefore
cannot
address
that
comment
specifically.

E.
Compression
Ignition
I.
E.
1
Comment:
One
commenter
(
526)
requested
that
EPA
delay
the
portion
of
the
rule
dealing
with
CI
engines
to
avoid
potential
issues
with
diesel
engines
in
California
and
some
other
states
being
subject
to
two
different
regulatory
regimes,
one
for
particulate
matter
(
PM)
and
the
other
for
formaldehyde,
thereby
requiring
two
different
control
technologies.

Response:
The
EPA
cannot
delay
the
portion
of
the
rule
dealing
with
CI
engines
because
it
is
under
statutory
and
court
deadlines
to
promulgate
standards
under
CAA
section
112
for
these
engines.
Also,
section
112
of
the
CAA
specifies
the
deadlines
required
for
meeting
standards
once
they
are
promulgated.
However,
EPA
does
not
feel
that
its
provisions
will
generally
have
the
effects
that
the
commenter
discusses.
The
rule
applies
only
to
new
CI
engines,
so
existing
CI
engines
will
not
be
affected.
Moreover,
for
those
sources
that
may
be
subject
to
California's
regulations
on
existing
engines,
they
should
be
able
to
deal
with
such
regulations
by
purchasing
engines
or
technologies
that
comply
with
both
requirements,
thus
reducing
some
costs.
Unfortunately,
EPA
is
not
able
to
stop
or
delay
the
rule
because
of
the
possible
plans
of
other
regulatory
agencies;
nor
can
EPA
stop
or
delay
regulation
of
one
type
of
pollutant
because
another
agency
may
be
regulating
a
different
pollutant
that
the
source
also
emits.
Reduction
of
pollution
sometimes
requires
multiple
controls
to
deal
with
multiple
pollutant
types.

I.
E.
2
Comment:
One
commenter
(
562)
expressed
that
EPA
has
correctly
concluded
that
the
MACT
floor
for
CI
engines
is
no
emissions
reductions.
The
EPA
states
less
than
1
percent
of
the
existing
CI
engines
have
emission
control
equipment
installed.
Because
the
CAA
specifies
that
the
MACT
floor
is
to
be
set
at
the
average
emission
limitation
achieved
by
the
best
performing
12
percent
of
existing
sources,
EPA
must
conclude
that
the
MACT
floor
is
no
emissions
reductions.
Also,
EPA
has
correctly
concluded
that
the
potential
beyond­
the­
floor
MACT
alternatives,
oxidation
catalyst
and
fuel
switching,
are
inappropriate
for
existing
CI
engines.

Response:
The
EPA
acknowledges
the
commenter's
support.
This
approach
has
been
28
retained
in
the
final
rule.

I.
E.
3
Comment:
One
commenter
(
546)
recommended
that
the
rule
include
a
provision
to
exempt
diesel
engines
in
California
that
meet
the
requirements
of
California's
Stationary
Diesel
Engine
Airborne
Toxic
Control
Measure
(
ATCM).
The
commenter
also
argued
that
the
recordkeeping,
reporting,
monitoring
and
testing
requirements
of
the
rule
are
not
appropriate
for
diesel
engines
meeting
a
diesel
PM
emission
standard.
The
commenter
also
felt
that
the
annual
source­
testing
requirement
in
the
rule
is
not
necessary,
appropriate
or
cost
effective
for
engines
with
diesel
particulate
filters
(
DPF).

The
commenter
expressed
the
view
that
the
rule
is
not
health
protective
in
that
it
only
regulates
organic
gases
and
ignores
diesel
PM.
The
rule
should
allow
State
and
local
agencies
to
regulate
diesel
PM
as
a
more
health
protective
alternative
for
addressing
the
risk
from
diesel
exhaust.

Response:
The
EPA
disagrees
with
the
suggestion
to
exempt
diesel
engines
meeting
California's
Stationary
Diesel
Engine
ATCM.
The
EPA
cannot
exempt
sources
in
a
source
category
from
meeting
the
MACT
requirements.
The
EPA
understands
that
California
has
been
regulating
diesel
engines
separately
to
reduce
PM
from
diesel
engines.
Such
emissions
are
not
regulated
in
the
final
rule.
However,
the
fact
that
a
source
in
California
must
meet
the
requirements
of
the
ATCM
does
not
excuse
it
from
meeting
the
separate
Federal
requirements
of
CAA
section
112.
The
commenter's
reference
to
section
209
is
unavailing.
The
provision
that
authorizes
California
to
regulate
mobile
("
nonroad")
engines
is
irrelevant
to
EPA's
separate
obligation
under
CAA
section
112
to
regulate
stationary
engines.

The
specific
pollutants
covered
in
the
final
rule
are
distinguishable
from
what
was
regulated
in
the
California
Air
Resources
Board
(
CARB)
rule.
While
California
may
believe
diesel
PM
to
be
a
greater
health
concern,
CAA
section
112
does
not
permit
EPA
to
ignore
the
regulation
of
listed
HAP
in
order
to
accommodate
the
separate
regulation
of
other
pollutants.
Moreover,
the
commenter
does
not
contend
that
the
NESHAP
will
increase
levels
of
diesel
PM
from
engines
that
have
a
PM
filter.

Of
course,
any
source
that
can
meet
the
pollutant
concentration
requirements
mandated
in
the
final
rule
using
a
filter,
as
well
as
any
engine­
out
controls
necessary
or
appropriate,
need
not
employ
a
catalyst.
If
the
information
provided
in
the
comment
is
accurate,
a
source
could
in
fact
use
certain
types
of
particulate
filters
(
likely
a
catalyzed
filter
would
be
used)
to
meet
the
requirements
of
the
NESHAP.

The
commenter
does
not
explain
why
the
recordkeeping,
reporting,
monitoring
and
testing
requirements
of
the
proposed
rule
are
inappropriate.
Numerous
sources
in
numerous
categories
are
required
to
meet
the
requirements
of
different
standards
promulgated
by
different
authorities.
The
commenter
does
not
explain
why
this
requirement
is
fundamentally
different
from
others
in
that
respect.
The
EPA
cannot
suspend
the
requirement
to
show
compliance
with
its
regulations
merely
because
another
jurisdiction
has
required
sources
to
show
compliance
with
its
regulations.
29
No
source
in
California
has
requested
the
relief
suggested
by
CARB.

I.
E.
4
Comment:
One
commenter
(
533)
indicated
that
the
usage
of
oxidation
catalyst
is
not
recommended
in
context
with
CI
engines
burning
fuels
containing
sulfur.
Oxidation
catalysts
less
sensitive
to
sulfur
dioxide
(
SO
2)
deactivation
usually
oxidize
a
large
proportion
of
the
SO
2
to
sulfate
(
SO
3)
and
as
an
end
result
additional
sulfate
PM
(
small
particulate)
is
formed.
Another
consequence
of
sulfate
formulation
is
clogging,
deactivation,
etc.
of
the
catalyst.
The
commenter
proposed
that
EPA
take
a
similar
standpoint
with
compression
ignition
engines
operating
on
fuels
containing
sulfur
as
was
taken
on
RICE
burning
digester
and
landfill
gas.

Response:
The
EPA
consulted
catalyst
vendors
who
stated
that
oxidation
of
SO
2
to
acid
mist
can
occur
over
a
catalyst.
However,
SO
2
emissions
from
both
natural
gas
and
distillate
oil
fired
engines
are
very
low,
therefore
the
impact
on
PM
emissions
would
be
very
low.
In
addition,
the
catalyst
can
be
installed
at
a
location
where
the
temperature
and
the
residence
time
would
help
to
minimize
the
potential
for
SO
3
formation.
During
EPA's
testing
at
Colorado
State
University
(
CSU),
low
sulfur
fuel
was
used
and
only
a
minimal
increase
in
PM
emissions
was
detected
(
about
3
percent
on
average).
The
EPA
expects
that
if
higher
sulfur
fuel
is
used,
there
would
be
increased
PM
emissions.
However,
EPA
expects
that
owners
and
operators
will
use
low
sulfur
fuel
where
the
increase
in
PM
is
expected
to
be
small.
Low
sulfur
fuel
and
ultra
low
sulfur
fuel
are
currently
available
and
will
become
more
available
as
EPA's
regulations
for
on
highway
diesel
fuel,
and,
as
expected,
nonroad
diesel
fuel,
are
phased
in
over
the
next
several
years.

F.
Subcategories
I.
F.
1
Comment:
One
commenter
(
555)
recommended
that
EPA
evaluate
the
ability
of
engines
associated
with
military
tactical
equipment
to
comply
with
the
standard
and
the
associated
costs
and
cost
effectiveness
and
environmental
benefits,
and
a
separate
category
for
these
engines
be
established
if
warranted.
Engines
associated
with
military
tactical
support
equipment
are
not
addressed
in
the
proposal.
These
engines
are
often
located
at
one
location
(
military
base)
for
more
than
12
months
and
hence
may
not
qualify
as
nonroad
engines.
However,
such
engines
often
have
unique
requirements
for
deployability
that
may
preclude
the
use
of
add­
on
air
pollution
control
equipment.

Response:
The
EPA
contacted
the
commenter
for
additional
information
regarding
this
comment.
Based
on
the
information
received,
it
appears
that
most
of
the
engines
the
commenter
is
concerned
with
are
typically
small
units.
The
EPA
would
expect
that
few
of
the
engines
in
question
would
be
greater
than
500
HP
in
size.
The
final
rule
does
not
cover
engines
500
HP
in
size
or
less.
For
the
engines
that
are
covered
by
the
final
rule,
the
engines
located
at
one
location
for
more
than
12
months
are
likely
being
stored
and
not
operated.
If
the
engines
do
not
operate,
they
do
not
have
to
comply
with
any
specific
requirements
of
the
final
rule.
When
these
engines
are
deployed
in
the
field
and
operate,
they
would
be
considered
mobile
sources
and
would
not
covered
by
the
final
rule
for
stationary
RICE.
For
these
reasons,
EPA
feels
that
no
separate
subcategory
is
needed
for
these
engines.
30
I.
F.
2
Comment:
One
commenter
(
555)
recommended
that
EPA
evaluate
the
ability
of
cyclic
engines
to
comply
with
the
standard
and
the
associated
costs
and
cost
effectiveness
and
environmental
benefits,
and
a
separate
category
for
engines
with
cyclic
operations
be
established
if
warranted.
The
test
results
upon
which
the
standards
are
based
do
not
cover
operations
at
less
than
50
percent
load.
Engines
in
cyclic
operations
may
spend
a
large
part
of
their
time
idling
or
operating
far
below
50
percent
load
with
only
brief
periods
at
high
load.
Since
the
control
effectiveness
for
oxidation
catalysts
and
NSCR
depends
on
the
exhaust
temperature
and
exhaust
temperature
is
expected
to
decrease
at
low
loads,
engines
with
cyclic
operations
may
not
be
able
to
comply
with
the
standards
or
be
able
to
comply
only
with
prohibitively
expensive
control
systems.

Response:
The
EPA
has
specified
in
the
final
rule
that
performance
tests
must
be
conducted
at
high
load
conditions,
defined
as
100
percent
±
10
percent.
Sources
are
not
required
to
meet
the
emission
limitations
at
low
load.

I.
F.
3
Comment:
One
commenter
(
577)
urged
EPA
to
consider
establishing
an
exception
for
hospital­
based
units
that
occasionally
peak
shave
in
response
to
excessively
high
spot
market
prices.
Only
a
few
hours
of
peak
shaving
can
provide
tremendous
financial
relief
from
highenergy
prices
while
producing
a
relatively
small
contribution
to
HAP.
An
American
Society
for
Health
Care
Engineering
poll
indicated
that
only
39
of
482
hospitals
routinely
or
occasionally
peak
shave
with
a
range
of
121­
161
hrs/
yr
(
in
addition
to
emergency
operation).
This
practice
falls
well
below
the
10
percent
utilization
in
the
proposed
limited
use
subcategory.

Response:
In
the
final
rule,
EPA
has
defined
limited
use
stationary
RICE
as
stationary
RICE
that
operate
less
than
100
hours
per
year.
There
is
no
limit
on
the
amount
of
hours
an
emergency
stationary
RICE
is
allowed
to
operate
for
emergency
purposes.
Emergency
stationary
RICE
may
use
an
additional
50
hours
per
year
for
non­
emergency
use.
The
EPA
feels
that
allowing
50
hours
per
year
is
appropriate
and
a
sufficient
amount
to
assist
owners
of
emergency
units.

I.
F.
4
Comment:
One
commenter
(
570)
requested
a
subcategory
for
new
and
reconstructed
stationary
CI
RICE
located
in
the
State
of
Alaska
that
exempts
the
engines
from
the
control
requirements
of
this
proposed
rule.
The
commenter
stated
that
EPA
has
overlooked
the
fact
that
low
sulfur
fuels
(
less
than
500
ppm
(
parts
per
million)
(
0.05
weight
percent))
are
necessary
for
CO
oxidation
catalysts
to
operate
properly
and
that
these
fuels
are
not
available
in
several
areas
of
the
United
States
including
the
State
of
Alaska.
Sulfur
can
quickly
degrade
oxidation
catalyst
performance
for
controlling
CO
(
or
formaldehyde)
emissions
by
poisoning
the
precious
metal
substrate
of
the
catalyst.
In
one
study
it
was
found
that
increasing
the
diesel
sulfur
content
from
3
ppm
to
350
ppm
by
weight
resulted
in
a
three­
fold
increase
in
catalyst­
out
PM
emissions.
In
the
same
study,
the
performance
of
the
diesel
oxidation
catalyst
for
controlling
CO
emissions
from
the
higher
sulfur
fuel
degraded
by
an
average
of
10
percent
after
the
short­
term
(
250­
hour)
aging
tests.
In
Alaska
meeting
the
proposed
MACT
floor
(
oxidation
catalyst)
for
new
CI
RICE
sources
will
be
problematic
because
of
the
non­
availability
of
low
sulfur
diesel
fuels
(
300
31
to
500
ppm
sulfur
content
by
weight).
The
permitted
diesel
fuel
sulfur
content,
by
weight,
for
most
permitted
stationary
CI
sources
is
between
0.1
percent
and
0.5
percent
(
1,000
ppm
to
5,000
ppm
by
weight).
The
Trans
Alaska
Pipeline
System
facilities
operated
by
the
commenter
have
permitted
sulfur
fuel
content
limits
between
0.24
percent
to
0.5
percent.
The
lowest
fuel
sulfur
diesel
that
is
available
in
the
State
of
Alaska
is
an
arctic
grade
fuel
that
has
a
sulfur
content
of
approximately
0.1
percent.
Petroleum
refineries
in
the
State
are
not
required
to
produce
lower
sulfur
fuels
because
Alaska
is
exempted
(
see
40
CFR
part
69
of
69
FR
34126)
from
EPA's
low
sulfur
highway
diesel
fuel
standards.

Response:
The
EPA
feels
it
is
unnecessary
to
establish
a
subcategory
for
new
and
reconstructed
CI
RICE
located
in
the
State
of
Alaska.
Information
received
from
the
Alaska
Department
of
Environmental
Conservation
(
DEC)
indicated
that
there
is
a
refinery
in
Alaska
that
can
produce
low
sulfur
fuel
(
300
to
500
ppm
sulfur
content
by
weight).
The
refinery
can
make
low
sulfur
diesel
that
meets
arctic
pour
point
specifications.
The
information
from
the
Alaska
DEC
also
indicated
that
low
sulfur
fuel
is
generally
available
where
there
are
roads
in
Anchorage,
but
not
generally
available
on
other
parts
of
the
road
system,
such
as
Fairbanks.
Some
remote
villages
do
have
low
sulfur
fuel.
The
EPA
expects
availability
to
grow
further
as
EPA's
final
rule
implementing
new
sulfur
limits
for
highway
fuel,
including
fuel
in
Alaska
(
68
FR
5002,
January
18,
2001),
is
implemented
beginning
in
2006.
The
Alaska
DEC
said
that
Alaska
has
200
small
villages
that
are
remote,
and
it
may
be
difficult
for
these
small
villages
to
always
have
low
sulfur
fuel
available.
These
villages
tend
to
employ
RICE
to
generate
electricity
and
have
between
two
to
four
stationary
RICE
in
their
power
plants.
These
engines
range
from
6
to
4000
kilowatt
(
kW),
with
an
average
of
300
kW.
The
Alaska
DEC
said
that
these
engines
are
below
the
threshold
for
major
sources,
and
that
is
also
confirmed
by
HAP
emission
calculations.
Since
these
villages
would
not
be
major
HAP
sites
they
would
not
be
affected
by
the
final
rule.
The
nonavailability
of
low
sulfur
fuel
at
these
remote
villages
would
therefore
not
be
an
issue
since
these
villages
would
not
be
subject
to
the
rule
since
they
are
located
at
non­
major
HAP
sites.
Finally,
EPA
has
received
information
from
catalyst
vendors
indicating
that
there
are
sulfur
tolerant
catalysts
that
have
been
commercialized
and
are
suitable
for
use
with
fuels
having
a
sulfur
content
between
3,000
and
5,000
ppm
sulfur
by
weight.
Sources
that
may
not
be
able
to
obtain
low
sulfur
fuel
could
use
such
catalysts
to
comply
with
the
requirements
of
the
final
rule.
For
these
reasons,
EPA
does
not
feel
it
is
necessary
to
establish
a
separate
subcategory
for
stationary
CI
RICE
located
in
Alaska.

I.
F.
5
Comment:
One
commenter
(
553)
expressed
strong
support
for
EPA's
proposal
to
create
a
subcategory
specifically
for
CI
engines.

Response:
The
EPA
recognizes
the
commenter's
support.
The
subcategory
for
CI
engines
has
been
retained
in
the
final
rule.

G.
Other
I.
G.
1
Comment:
One
commenter
(
586)
expressed
the
opinion
that
the
health
risks
of
32
uncontrolled
engines
are
sufficient
to
warrant
consideration
of
making
this
MACT
standard
applicable
to
non­
major
HAP
facilities
and
other
types
of
engines.

Response:
The
EPA
is
covering
area
sources
and
small
engines
under
a
later
rule,
which
is
expected
to
be
proposed
by
October
31,
2006
and
completed
by
December
20,
2007.

I.
G.
2
Comment:
One
commenter
(
586)
stated
that
the
proposed
MACT
should
also
apply
to
existing
lean
burn
engines,
not
only
existing
rich
burn
engines.
Retrofitting
an
oxidation
catalyst
on
a
lean
burn
engine
should
not
be
much
more
complicated
than
retrofitting
a
nonselective
catalyst
to
a
rich
burn
engine.

Response:
As
established
in
section
112(
d)
of
the
CAA,
the
emission
standards
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.
Review
of
the
Population
Database
indicated
that
less
than
1
percent
of
existing
2SLB
stationary
RICE,
about
3
percent
of
existing
4SLB
stationary
RICE,
and
less
than
1
percent
of
existing
CI
stationary
RICE
use
oxidation
catalyst
systems
or
other
emission
reduction
technologies
or
strategies.
These
percentages
are
all
well
below
the
criteria
for
a
MACT
floor
that
would
require
use
of
such
systems,
or
equivalent
emissions
reductions
for
existing
sources.
The
EPA
concluded
that
the
MACT
floor
for
existing
2SLB,
existing
4SLB,
and
existing
CI
stationary
RICE
is
no
emissions
reductions.
Unlike
the
case
for
lean
burn
and
CI
engines,
the
EPA
determined
that
27
percent
of
existing
4SRB
stationary
RICE
are
equipped
with
NSCR.
Therefore
it
was
appropriate
to
set
the
MACT
floor
for
4SRB
stationary
RICE
at
the
level
of
HAP
emissions
reductions
achieved
by
the
use
of
NSCR
systems.
As
discussed
in
the
proposed
rule,
EPA
also
determined
that
it
was
not
appropriate
to
require
above­
the­
floor
emissions
reductions
for
the
subcategories
in
this
source
category.

I.
G.
3
Comment:
One
commenter
(
565)
requested
that
EPA
clarify
applicability
and
exemptions
with
a
flowchart.

Response:
The
EPA
feels
that
the
manner
in
which
the
applicability
and
exemptions
of
the
rule
are
presented
is
sufficient.

I.
G.
4
Comment:
One
commenter
(
585)
requested
clarification
on
the
requirements
for
dual­
fuel
engines.
Reciprocating
internal
combustion
engines
are
sometimes
modified
to
use
fuels
such
as
diesel
and
natural
gas.
The
dual
fuels
often
are
5
percent
diesel
and
95
percent
natural
gas.
Some
clean
burn
dual
RICE
units
use
only
1
percent
diesel
and
99
percent
natural
gas.
It
is
not
clear
how
these
engines
should
be
defined
for
compliance.
The
commenter
stated
that
it
would
be
better
to
define
the
RICE
units
according
to
ignition
instead
of
fuel.
The
commenter
suggested
the
following
wording
concerning
dual­
fuel
RICE
units:
"
Engines
that
are
modified
to
reduce
emission
will
be
categorized
or
will
continue
to
be
categorized
as
natural
gas
engines
if
the
ignition
system
is
typical
of
a
natural
gas
engine,
i.
e.
spark
ignition
or
diesel
engine
if
the
ignition
system
is
typical
of
diesel
engines,
i.
e.
compression
ignition."
33
One
commenter
(
555)
suggested
that
EPA
evaluate
the
ability
of
dual­
fuel
engines
to
comply
with
the
standard
and
the
associated
costs
and
cost
effectiveness
and
environmental
benefits,
and
a
separate
category
for
these
engines
be
established
if
warranted.

Response:
In
the
rule,
EPA
defined
CI
engine
as
any
stationary
RICE
"
in
which
a
high
boiling
point
liquid
fuel
injected
into
the
combustion
chamber
ignites
when
the
air
charge
has
been
compressed
to
a
temperature
sufficiently
high
for
auto­
ignition,
including
diesel
engines
and
dualfuel
engines
and
engines
that
are
not
spark
ignition."
Dual­
fuel
engines
in
turn
are
defined
as
engines
"
in
which
a
liquid
fuel
(
typically
diesel
fuel)
is
used
for
compression
ignition
and
gaseous
fuel
(
typically
natural
gas)
is
used
as
the
primary
fuel."
Therefore,
dual­
fuel
units
as
defined
are
required
to
comply
with
the
emission
limitation
requirements
established
in
the
final
rule
for
CI
engines.

In
response
to
the
commenter
who
stated
that
EPA
should
evaluate
the
ability
of
dual­
fuel
engines
to
comply
with
the
standard,
EPA
does
not
have
any
reason
to
expect
that
dual­
fuel
engines
will
not
be
able
to
meet
the
standards
of
the
final
rule.
Information
contained
in
several
papers
available
from
EPA's
Edocket
Website
as
attachments
to
Document
ID
Number
OAR­
2002­
0059­
0503
(
A­
95­
35
II­
D­
78)
shows
that
dual­
fuel
engines
equipped
with
oxidation
catalysts
can
achieve
CO
removal
efficiencies
of
93
percent
or
more.
Since
the
emission
limitations
EPA
is
requiring
in
the
final
rule
range
from
58
to
93
CO
percent
reduction,
dual­
fuel
engines
should
be
able
to
meet
the
CO
percent
reduction
emission
limitations.

I.
G.
5
Comment:
One
commenter
(
569)
provided
support
for
EPA's
determination
to
set
the
MACT
standard
at
the
MACT
floor.
The
commenter
supported
this
approach
for
the
following
reasons:

(
1)
The
standard
can
be
achieved
by
most
existing
and
new
sources
in
the
source
category
in
a
cost
effective
manner;
and
(
2)
The
standard
will
provide
a
sufficient
level
of
emission
controls
to
adequately
protect
the
public
and
the
environment.

Response:
The
EPA
acknowledges
the
commenter's
support.
This
approach
has
been
retained
in
the
final
rule.

II.
DEFINITIONS
A.
Rich/
Lean
Burn
II.
A.
1
Comment:
Six
commenters
(
542,
549,
554,
556,
565,
587)
stated
that
EPA
should
revise
the
definition
of
rich
burn
engine
to
eliminate
engines
that
have
been
converted
to
operate
as
lean
burn
engines
and
to
address
older
engines
(
e.
g.,
horizontal),
where
there
is
no
recommended
air/
fuel
ratio.
Commenter
587
recommended
that
EPA
adopt
the
following
34
definition
into
the
final
rule:

"
Rich
burn
engine
means
4­
stroke
spark­
ignited
engine
where
the
manufacturers
recommended
air/
fuel
ratio
divided
by
the
stoichiometric
air/
fuel
ratio
at
full
conditions
is
less
than
or
equal
to
1.1.
Engines
originally
manufactured
as
rich­
burn
engines,
but
modified
prior
to
(
promulgation
date)
with
passive
emission
control
technology
for
nitrogen
dioxides
(
such
as
precombustion
chambers)
shall
be
considered
lean
burn
engines.
Horizontal
engines
shall
be
considered
lean
burn
engines.
Also,
older
engines
where
there
are
no
manufacturer's
recommendations
regarding
air/
fuel
ratio
will
be
considered
a
rich­
burn
engine
if
the
excess
oxygen
content
of
the
exhaust
at
full
load
conditions
is
less
than
or
equal
to
2
percent."

Commenter
549
recommended
a
similar
definition.

Response:
The
EPA
agrees
with
the
commenter
that
it
is
necessary
to
address
engines
that
have
been
converted
from
4SRB
engines
to
4SLB
engines
and
to
also
address
older
engines
such
as
horizontal
engines.
The
EPA
has
adjusted
the
definition
of
rich
burn
engine
and
has
written
the
rich
burn
definition
in
the
final
rule
as
follows:

"
Rich
burn
engine
means
any
four­
stroke
spark
ignited
engine
where
the
manufacturer's
recommended
operating
air/
fuel
ratio
divided
by
the
stoichiometric
air/
fuel
ratio
at
full
load
conditions
is
less
than
or
equal
to
1.1.
Engines
originally
manufactured
as
rich
burn
engines,
but
modified
prior
to
December
19,
2002
with
passive
emission
control
technology
for
NOx
(
such
as
pre­
combustion
chambers)
will
be
considered
lean
burn
engines.
Also,
existing
engines
where
there
are
no
manufacturer's
recommendations
regarding
air/
fuel
ratio
will
be
considered
a
rich
burn
engine
if
the
excess
oxygen
content
of
the
exhaust
at
full
load
conditions
is
less
than
or
equal
to
2
percent."

In
addition,
to
avoid
overlap
with
the
definition
of
lean
burn
engine,
the
lean
burn
engine
definition
has
also
been
adjusted
and
reads
as
follows
in
the
final
rule:

"
Lean
burn
engine
means
any
two­
stroke
or
four­
stroke
spark
ignited
engine
that
does
not
meet
the
definition
of
a
rich
burn
engine."

II.
A.
2
Comment:
One
commenter
(
552)
suggested
that
EPA
revise
the
definition
of
rich
burn
engine
to
"
A
spark­
ignited
engine
that
is
designed
to
operate
with
an
exhaust
oxygen
concentration
of
less
than
2
percent
by
dry
volume."
The
commenter
was
of
the
opinion
that
the
current
definition
does
not
necessarily
include
all
engines
considered
to
be
rich
burn.
In
addition,
according
to
the
commenter,
determining
whether
an
engines
falls
into
the
rich­
burn
category
under
the
current
definition
is
problematic
since
obtaining
the
data
to
calculate
the
ratio
stated
in
the
rule
is
very
difficult.

Response:
As
described
in
response
to
comment
II.
A.
1,
EPA
has
adjusted
the
definition
of
rich
burn
engine
in
the
final
rule.
The
EPA
feels
the
definition
of
rich
burn
engine
in
the
final
rule
satisfies
the
commenter's
concerns.
35
II.
A.
3
Comment:
One
commenter
(
527)
observed
that
in
the
proposed
rule
there
is
no
mention
if
a
rich
to
lean
conversion
of
a
4SRB
engine
is
permitted.
The
commenter
indicated
that
such
a
conversion
should
be
permitted
due
to
economic
and
environmental
factors.

Response:
The
definition
of
a
rich
burn
engine
is
based
on
the
manufacturer's
recommendation.
Therefore,
a
conversion
of
a
4SRB
engine
to
a
4SLB
engine
would
not
change
the
classification
of
the
engine
as
a
4SRB
engine,
unless
the
engine
was
originally
manufactured
as
a
rich
burn
engine,
but
modified
prior
to
December
19,
2002
with
passive
emission
control
technology
for
nitrogen
oxides
(
NOx)
(
such
as
pre­
combustion
chambers).

II.
A.
4
Comment:
One
commenter
(
600)
recommended
that
in
order
to
eliminate
the
ambiguity
that
a
CI
RICE
could
be
classified
as
a
2SLB
or
4SLB
RICE,
the
"
lean
burn
engine"
definition
should
be
changed
to
reflect
that
it
should
be
applicable
only
to
spark
ignited
engines.
The
commenter
recommended
the
following
change:
"
Lean
burn
engine
means
any
two­
stroke
or
four­
stroke
spark
ignited
engine
where
the
manufacturer's
recommended
operating
air/
fuel
ratio
divided
by
the
stoichiometric
air/
fuel
ratio
is
greater
than
1.1."
The
proposed
change
is
consistent
with
the
language
in
the
preamble
where
2SLB
and
4SLB
are
often
preceded
by
either
of
the
terms
"
spark
ignited"
or
"
spark
ignition."

Response:
The
EPA
agrees
with
the
commenter
and
has
incorporated
parts
of
the
commenter's
recommendation
into
the
definition
of
lean
burn
engine
in
the
final
rule.
The
definition
of
lean
burn
engine
in
the
final
rule
is
written
as
follows:

"
Lean
burn
engine
means
any
two­
stroke
or
four­
stroke
spark
ignited
engine
that
does
not
meet
the
definition
of
a
rich
burn
engine."

B.
New/
Reconstructed
II.
B.
1
Comment:
One
commenter
(
546)
asserted
that
the
definition
of
a
reconstructed
source
should
be
modified
to
exclude
any
cost
incurred
with
the
installation
of
a
control
device
required
by
State
and
local
emission
standards.
The
addition
of
DPF
could
exceed
the
reconstruction
cost
threshold
(
50
percent
of
fixed
capital
cost
to
construct
a
comparable
new
source).

Response:
Based
on
the
information
EPA
has
available
on
costs
of
DPF
systems
and
costs
of
engines,
EPA
feels
that
the
addition
of
DPF
would
not
exceed
the
reconstruction
threshold
of
50
percent
of
the
capital
cost
of
a
new
engine.
Information
received
from
CARB
indicate
that
the
total
cost
of
a
DPF
including
equipment
and
installation
is
around
$
38/
HP.
Engine
costs
estimated
by
CARB
are
$
93/
HP
for
a
new
engine.
Comparing
the
cost
of
a
DPF
system
to
the
cost
of
a
new
engine
shows
that
the
addition
of
such
a
filter
system
would
be
less
than
50
percent.
Engine
cost
information
available
to
EPA
obtained
from
other
sources
indicate
that
engine
costs
are
between
$
150­$
270/
HP.
Using
these
engine
costs,
the
addition
of
a
DPF
system
would
be
an
even
less
percentage
of
the
cost
of
a
new
engine.
Engine
costs
can
be
found
in
the
memorandum
entitled
"
National
Impacts
Associated
with
Reciprocating
Internal
36
Combustion
Engines"
available
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0071
(
A­
95­
35
II­
B­
48).
The
EPA
has,
therefore
concluded
that
based
on
both
information
received
from
CARB
and
information
EPA
already
has,
the
addition
of
a
DPF
would
be
less
than
50
percent
of
the
cost
of
a
new
engine.

In
any
case,
EPA's
policy
regarding
the
inclusion
of
air
pollution
control
equipment
in
determining
reconstruction
is
that
the
costs
associated
with
the
purchase
and
installation
of
air
pollution
control
equipment
are
included
in
the
fixed
capital
cost
to
the
extent
that
the
equipment
is
required
as
part
of
the
manufacturing
or
operating
process.
Therefore,
it
is
EPA's
policy
not
to
include
the
fixed
capital
cost
of
air
pollution
control
equipment
that
is
not
part
of
the
operating
process.
Since
DPF
is
not
required
in
order
to
operate
an
engine,
the
cost
for
purchase
and
installation
of
DPF
would
not
be
included
in
determining
whether
a
source
is
reconstructed.
The
commenter
does
not
explain
why
EPA
should
deviate
from
its
General
Provisions
based
on
compliance
with
State
or
local
regulations.
A
source
that
is
spending
more
than
50
percent
of
the
capital
cost
needed
for
a
new
engine
to
meet
the
requirements
should
be
in
a
position
to
make
appropriate
changes
in
its
source
at
that
time
to
meet
the
standards
promulgated
today.
Moreover,
the
source
may
be
able
to
comply
with
both
requirements
at
the
same
time
and
may
be
able
to
meet
the
requirements
using
integrated
controls
(
if
not
the
same
controls)
that
would
be
best
implemented
at
the
same
time.

II.
B.
2
Comment:
One
commenter
(
552)
recommended
that
EPA
change
the
definition
of
new
and
reconstructed
RICE
to
include
those
engines
where
construction
or
reconstruction
is
initiated
after
the
date
of
publication
of
the
final
rule
in
the
Federal
Register.
The
commenter
stated
that
this
requirement
is
unreasonable
since
it
places
an
unfair
burden
on
owners
who
install
RICE
in
the
time
period
between
publication
of
the
draft
rule
and
publication
of
the
final
rule.
One
issue
the
commenter
raised
was
that
since
the
final
standards
and
requirements
will
not
be
known
until
the
final
publication
data,
any
new
RICE
installed
since
December
19,
2002,
could
not
possibly
be
in
compliance
on
the
final
publication
date
and
will
have
to
shut
down
operations
until
compliance
with
all
requirements
can
be
assured.

Response:
The
EPA
feels
the
definitions
of
new
and
reconstructed
RICE
as
defined
in
§
§
63.6590(
a)(
2)
and
(
3)
of
the
final
rule
are
appropriate.
These
definitions
are
also
consistent
with
the
General
Provisions
of
part
63
and
with
the
definitions
in
CAA
section
112(
a).
Sources
are
required
to
install
the
proper
equipment
and
meet
the
applicable
emission
limitations
on
startup;
however,
EPA
allows
sources
180
days
to
demonstrate
compliance.
In
addition,
if
the
promulgated
standard
is
more
stringent
that
the
proposed
standard,
sources
that
commenced
construction
or
reconstruction
between
proposal
and
promulgation
have
the
option
to
comply
with
the
proposed
emission
limitations
when
demonstrating
initial
compliance.
Sources
must
then
meet
the
final
emission
limitation
within
3
years
of
the
final
rule
and
must
conduct
a
second
performance
test
to
demonstrate
compliance
with
the
promulgated
emission
limitations
within
3
years
and
180
days
after
the
date
of
the
final
rule.

II.
B.
3
Comment:
One
commenter
(
553)
stated
that
it
strongly
supports
EPA's
definition
of
"
reconstructed"
source.
37
Response:
The
EPA
acknowledges
the
commenter's
support.

C.
Affected
Source
II.
C.
1
Comment:
Nine
commenters
(
544,
549,
551,
553,
564,
565,
581,
585,
593)
requested
that
EPA
write
the
definitions
of
affected
source,
existing
stationary
RICE,
new
stationary
RICE,
and
reconstructed
stationary
RICE
such
that
they
represent
the
"
collection"
of
each
type
of
source
at
a
site,
consistent
with
General
Provisions
§
63.2.

Response:
Although
§
63.2
of
the
General
Provisions
provides
that
EPA
will
generally
adopt
a
broad
definition
of
affected
source,
which
includes
all
emission
units
within
each
subcategory
which
are
located
within
the
same
contiguous
area,
this
section
also
provides
that
EPA
may
adopt
a
narrower
definition
of
affected
source
in
instances
where
EPA
determines
that
the
broader
definition
would
"
create
significant
administrative,
practical,
or
implementation
problems"
and
"
the
different
definition
would
resolve
those
problems."
This
is
such
an
instance.
There
are
several
subcategories
of
stationary
RICE,
and
a
site
could
have
engines
from
multiple
subcategories,
each
having
different
compliance
requirements.
Use
of
the
broader
definition
of
affected
source
specified
by
the
General
Provisions
would
require
very
complex
aggregate
compliance
determinations.
The
EPA
feels
such
complicated
compliance
determinations
to
be
impractical,
and,
therefore,
has
decided
to
adopt
a
definition
which
establishes
each
individual
RICE
as
the
affected
source.

II.
C.
2
Comment:
One
commenter
(
551)
asked
that
EPA
clarify
the
term
"
affected
RICE"
e.
g.,
used
in
§
63.6595(
c).
The
commenter
stated
that
the
term
should
be
changed
to
"
affected
source"
or
"
any
stationary
RICE
that
is
part
of
an
affected
source"
or
"
any
stationary
RICE
subject
to
this
provision,"
as
appropriate
to
the
intended
context.

Response:
The
EPA
agrees
with
the
commenter
and
has
replaced
"
affected
RICE"
with
"
affected
source"
in
§
63.6595(
c)
of
the
final
rule.

D.
Other
II.
D.
1
Comment:
One
commenter
(
564)
contended
that
the
1­
year
limitation
on
engines
that
are
portable
or
transportable
before
they
are
categorized
as
stationary
should
be
eliminated.
For
purpose
of
the
RICE
MACT,
adding
on
control
requirements,
and
the
associated
control
systems,
to
engines
that
are
portable
is
not
practical.

Response:
The
commenter
did
not
provide
any
information
supporting
why
the
addition
of
control
equipment
to
portable
and
transportable
engines
that
have
remained
at
the
same
location
for
more
than
1
year
and
that
have
become
classified
as
stationary
engines
is
not
practical.
Therefore,
the
EPA
cannot
justify
eliminating
the
1­
year
limitation
without
a
reasonable
basis.

II.
D.
2
Comment:
One
commenter
(
555)
said
that
"
Gaseous
fuel"
should
be
defined
as
38
being
a
gas
at
a
standard
temperature
and
pressure
rather
than
"
normally
a
gas."

Response:
The
EPA
agrees
with
the
commenter
and
has
defined
gaseous
fuel
according
to
the
commenter's
recommendation
in
the
final
rule.

II.
D.
3
Comment:
One
commenter
(
555)
questioned
the
definition
of
"
Landfill
gas"
and
stated
that
"
Landfill
gas"
is
defined
as
being
"
formed
through
anaerobic
decomposition,"
which
is
usually
but
not
always
the
case.

Response:
The
definitions
of
landfill
gas
and
digester
gas
have
been
modified
in
the
final
rule
to
indicate
that
they
are
typically
formed
through
anaerobic
decomposition.

II.
D.
4
Comment:
One
commenter
(
555)
remarked
that
"
Oxidation
catalyst"
is
defined
as
"
for
lean
burn
engines"
even
though
they
can
be
used
on
rich
burn
engines
with
air
injection.

Response:
The
EPA
has
adjusted
the
definition
of
oxidation
catalyst
in
the
final
rule
by
removing
"
for
lean
burn
engines."

II.
D.
5
Comment:
One
commenter
(
555)
provided
the
comment
that
"
Propane"
is
defined
as
"
suitable
for
use
in
spark­
ignited
engines"
when
it
could
be
used
in
a
dual­
fueled
engine
as
well.

Response:
The
EPA
agrees
with
the
commenter.
The
definition
of
propane
has
been
adjusted
to
exclude
the
sentence
"
suitable
for
use
in
spark­
ignited
engines"
in
the
final
rule.

II.
D.
6
Comment:
One
commenter
(
555)
made
the
observation
that
"
Nameplate
capacity"
is
not
defined
in
the
proposed
rule.
The
commenter
recommended
that
the
term
"
manufacturers
rating
at
maximum
continuous
load"
be
used.

Response:
The
EPA
received
several
comments
on
the
proposed
rule
stating
that
siterated
HP
should
be
used
instead
of
manufacturer's
nameplate
HP.
This
was
discussed
in
response
to
comment
X.
B.
1.
For
the
reasons
provided
by
EPA
in
response
to
X.
B.
1,
EPA
feels
it
is
appropriate
to
used
site­
rated
HP
as
the
basis
for
the
size
threshold
in
the
final
rule.

II.
D.
7
Comment:
One
commenter
(
564)
recommended
that
the
preamble
should
clarify
that
the
definition
of
major
source
in
the
RICE
MACT
does
not
alter
the
definition
of
a
major
source
in
subpart
HH
of
40
CFR
part
63
(
Oil
and
Natural
Gas
Production
Facilities)
and,
therefore,
does
not
affect
subpart
HH
applicability.
Similarly,
commenter
549
stated
that
EPA
should
clarify
in
the
preamble
that
major
source
status
is
MACT
category
specific
and
that
area
and
major
source
determinations
do
not
transfer
between
MACT
categories.

Response:
The
EPA
recognizes
the
commenter's
concern
regarding
the
definition
of
major
source
in
the
RICE
NESHAP
and
its
difference
from
the
definition
of
major
source
in
subpart
HH.
The
EPA
has,
therefore,
clarified
in
the
preamble
to
the
final
rule
that
the
definition
of
major
source
in
the
RICE
NESHAP
does
not
alter
the
definition
of
major
source
in
subpart
HH
39
(
or
any
other
subpart),
and,
therefore,
does
not
affect
subpart
HH
applicability.

II.
D.
8
Comment:
Two
commenters
(
564,
565)
stated
that
the
term,
and
definition
of,
"
deviation"
should
be
changed
to
"
excursion"
to
avoid
confusion
with
title
V
operating
permit
requirements
and
for
consistency
with
other
MACTs
(
e.
g.,
Hazardous
Organic
NESHAP).

Response:
The
EPA
does
not
agree
with
the
commenter
that
the
definition
of
deviation
should
be
changed
to
excursion.
The
term
"
deviation"
has
commonly
been
used
in
other
MACT
standards.
The
EPA
has
therefore
not
made
this
change.

II.
D.
9
Comment:
One
commenter
(
604)
stated
that
the
standard
exempts
RICE
whose
primary
fuel
is
landfill
gas
because
catalytic
converters
will
be
rendered
useless
by
siloxanes.
The
commenter
said
that
primary
fuel
is
not
defined
in
the
MACT.

Response:
The
EPA
agrees
that
it
is
appropriate
to
provide
guidelines
for
the
usage
of
landfill
and
digester
gas.
As
discussed
elsewhere
in
this
document,
firing
greater
than
10
percent
landfill
or
digester
gas
will
cause
fouling
that
will
render
an
oxidation
catalyst
inoperable
within
a
short
period
of
time.
Accordingly,
EPA
has
defined
this
subcategory
as
engines
with
annual
landfill
and
digester
gas
consumption
of
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis
in
the
final
rule.

II.
D.
10
Comment:
One
commenter
(
549)
expressed
support
for
the
definition
of
major
source,
except
as
follows:

The
phrase
"
except
when
they
are
on
the
same
surface
site"
should
be
removed
from
the
end
of
part
(
1)
of
the
RICE
major
source
definition.
This
phrase
is
not
present
in
the
40
CFR
part
63,
subpart
HH
major
source
definition
that
is
the
template
for
the
RICE
major
source
definition.
Section
112(
n)(
4)
of
the
CAA
of
1990
requires
that
wells
and
associated
equipment
not
be
aggregated
even
within
the
same
surface
site
except
as
provided
in
part
(
3)
of
the
RICE
MACT
major
source
definition.

In
part
(
3)
of
the
RICE
major
source
definition,
the
phrase
"
storage
vessel
with
flash
emissions
potential"
should
be
changed
to
"
storage
vessel
with
the
potential
for
flash
emissions"
to
conform
to
the
40
CFR
part
63,
subpart
HH
definition.

The
General
Provisions
major
source
definition
presented
in
the
RICE
MACT
is
different
from
those
found
in
the
definition
of
major
source
in
the
NESHAP
from
Oil
and
Natural
Gas
Production
Facilities
(
40
CFR
63.761).
The
significance
of
this
difference
is
that
sources
that
are
area
sources
under
subpart
HH
could
possibly
be
rendered
"
Major
Sources"
under
RICE.
The
EPA
should
acknowledge
this
possibility
in
the
preamble
to
the
final
rule
and
clearly
state
that
this
does
not
change
the
source's
status
under
subpart
HH
or
any
other
MACT.

Response:
The
EPA
agrees
that
the
phrase
"
except
when
they
are
on
the
same
surface
site"
should
be
removed
from
the
end
of
part
(
1)
of
the
RICE
major
source
definition
and
the
40
phrase
"
storage
tanks
with
flash
emissions
potential"
in
part
(
3)
of
the
RICE
MACT
major
source
definition
should
be
changed
to
"
storage
vessel
with
the
potential
for
flash
emissions."
The
EPA
has
incorporated
the
commenter's
recommendations
into
the
final
rule.

The
EPA
recognizes
the
commenters'
concern
regarding
the
definition
of
major
source
in
the
RICE
NESHAP
and
its
difference
from
the
definition
of
major
source
in
subpart
HH.
The
preamble
to
the
final
rule
clarifies
that
the
definition
of
major
source
in
the
RICE
MACT
does
not
alter
the
definition
of
major
source
in
subpart
HH,
and
therefore
does
not
affect
subpart
HH
applicability.

II.
D.
11
Comment:
One
commenter
(
549)
recommended
that
the
definitions
from
40
CFR
subpart
HH
and
40
CFR
subpart
HHH
for
glycol
dehydration
unit,
storage
vessel
with
the
potential
for
flash
emissions,
and
production
well
should
be
included.

Response:
The
EPA
agrees
with
the
commenter
that
the
definitions
should
be
included
in
the
RICE
NESHAP.
The
definitions
from
40
CFR
subpart
HH
and
40
CFR
subpart
HHH
for
glycol
dehydration
unit,
storage
vessel
with
the
potential
for
flash
emissions,
and
production
well
have
been
added
to
the
final
RICE
NESHAP.

III.
DATES
A.
Comment
Period
III.
A.
1
Comment:
One
commenter
(
603(
B))
stated
that
a
60
day
comment
period
was
insufficient.

Response:
The
EPA
feels
the
comment
period
was
sufficient.

B.
New/
Reconstructed
III.
B.
1
Comment:
One
commenter
(
570)
stated
that
§
§
63.6595(
a)(
2)
and
(
3)
are
not
consistent
with
§
63.6590(
a)(
2).
"
New
stationary
RICE"
is
defined
as
any
RICE
for
which
construction
was
commenced
after
December
19,
2002
in
§
63.6590(
a)(
2),
while
§
63.6595(
a)(
2)
and
(
3)
indicate
"
new
stationary
RICE"
are
those
for
which
construction
commenced
after
the
final
publication
date
of
the
rule.
The
commenter
suggested
correcting
the
language
in
§
§
63.6595(
a)(
2)
and
(
3).
The
commenter
noted
that
the
problem
is
also
repeated
in
§
§
63.6645(
b)
and
(
c).

Response:
The
commenter's
interpretation
of
the
regulatory
text
is
not
accurate.
Section
63.6590(
a)(
2)
states
that
a
stationary
RICE
is
new
if
construction
of
the
stationary
RICE
was
commenced
on
or
after
December
19,
2002
(
date
of
publication
of
the
proposed
rule
in
the
Federal
Register).
Sections
63.6595(
a)(
2)
and
(
3)
do
not
define
new
stationary
RICE,
rather
they
state
when
the
new
or
reconstructed
RICE
must
comply
with
the
applicable
emission
limitations
41
depending
on
when
startup
occurred
(
before
or
after
publication
of
the
final
rule
in
the
Federal
Register).

C.
Initial
Notification
III.
C.
1
Comment:
Four
commenters
(
542,
554,
556,
587)
remarked
that
EPA
should
provide
1
year
for
initial
notification
as
in
the
glycol
dehydration
MACT.

Response:
An
initial
notification
is
not
a
time
consuming
activity,
and
the
EPA
does
not
feel
that
1
year
is
necessary
to
submit
an
initial
notification.

D.
Initial
Compliance
III.
D.
1
Comment:
Seven
commenters
(
526,
548,
552,
595,
560,
571,
579)
expressed
the
view
that
immediate
compliance
for
new
and
reconstructed
engines
is
unreasonable.
The
commenters
felt
that
1
year
compliance
time
frame
is
more
reasonable.

Response:
The
EPA
feels
that
immediate
compliance
is
appropriate
for
new
and
reconstructed
engines
and
is
consistent
with
the
General
Provisions
of
part
63.
See
also
CAA
section
112(
i)(
1).
The
requirements
of
CAA
section
112
contemplate
that
sources
will
be
aware
of
their
requirements
at
the
time
of
proposal
and,
excluding
requirements
that
are
made
more
stringent
between
proposal
and
promulgation,
new
or
reconstructed
sources
should
be
prepared
to
meet
such
requirements
immediately,
at
the
time
of
the
final
rule.
Sources
are
required
to
install
the
proper
equipment
and
meet
the
applicable
emission
limitations
on
startup;
however,
EPA
allows
sources
180
days
to
demonstrate
compliance.
In
addition,
because
two
of
our
emission
requirements
have
been
made
more
stringent
since
proposal,
sources
that
commence
operation
between
proposal
and
the
final
rule
may
show
compliance
with
the
proposed
requirements
for
the
first
3
years
of
the
program.

III.
D.
2
Comment:
One
commenter
(
569)
supported
the
3­
year
compliance
period
for
existing
affected
sources
provided
in
§
63.6595(
a)(
1)
because
this
amount
of
time
will
be
required
for
the
specification,
purchase,
manufacture,
installation
and
testing
of
the
required
catalyst
systems,
particularly
if
many
other
sources
are
doing
the
same
things
at
the
same
time.

Response:
The
EPA
agrees
with
the
commenter.

III.
D.
3
Comment:
Eight
commenters
(
526,
540,
559,
564,
565,
569,
579,
585)
stated
that
for
area
sources
becoming
major
sources,
the
requirement
to
be
in
compliance
at
the
time
of
the
switch
is
unreasonable.
Commenters
526
and
579
suggested
allowing
1
year
for
the
unit
to
come
into
compliance.
One
commenter
(
540)
suggested
that
all
area
sources
that
become
major
should
be
allowed
3
years
to
achieve
compliance
or
change
the
definition
of
a
new
stationary
RICE
to
"
A
stationary
RICE
is
new
if
you
commenced
construction
of
the
stationary
RICE
after
December
19,
2002,
and
you
meet
the
applicability
criteria
for
the
subpart
at
the
time
you
commenced
construction."
Six
commenters
(
549,
569,
559,
565,
564,
585)
suggested
3
years.
42
Response:
The
EPA
agrees
with
the
commenters
that
it
is
appropriate
to
allow
existing
area
sources
that
become
major
sources
3
years
to
comply
with
the
final
rule.
This
has
been
specified
in
the
final
rule
in
§
63.6595(
b)(
2).
However,
the
EPA
does
not
agree
with
the
commenters
that
immediate
compliance
is
unreasonable
for
new
and
reconstructed
RICE
located
at
area
sources
that
are
constructed
or
reconstructed
at
the
same
time
the
area
source
becomes
a
major
source.
These
sources
are
aware
in
advance
of
their
change
in
status
from
area
source
to
major
source
and,
therefore,
should
have
sufficient
time
to
plan
for
immediate
compliance
with
the
final
rule.
This
has
been
specified
in
the
final
rule
in
§
63.6595(
b)(
1).
A
period
of
180
days
is
allowed
to
demonstrate
compliance.

III.
D.
4
Comment:
Three
commenters
(
542,
556,
587)
requested
that
EPA
provide
1
year
to
conduct
the
initial
performance
test,
rather
than
180
days
provided
by
the
General
Provisions.
Commenter
587
indicated
that
seasonal
operations,
such
as
storage
facilities
or
compressor
stations
used
in
peak
demand
only,
may
not
be
operational
during
the
180
days
provided
to
conduct
the
performance
test.
All
existing
4SRB
engines
must
conduct
formaldehyde
testing
as
a
part
of
the
initial
performance
test.
It
may
be
difficult
to
secure
appropriate
testing
firms
within
the
180
days
provided,
especially
since
many
may
depend
on
Fourier
Transform
Infrared
(
FTIR)
testing.

Response:
The
EPA
feels
the
time
it
allows
sources
to
conduct
the
initial
performance
test
is
appropriate.
Existing
sources
that
must
meet
the
requirements
of
the
final
rule
have
3
years
and
180
days
to
conduct
the
initial
performance
test
and
to
demonstrate
compliance.
Therefore,
existing
4SRB
engines
that
must
meet
the
formaldehyde
emission
limitations
have
sufficient
time
to
secure
an
appropriate
testing
firm.
In
addition,
the
final
rule
does
not
only
specify
that
FTIR
can
be
used
for
formaldehyde
testing,
but
that
also
EPA
Method
323.
This
means
it
may
not
be
necessary
to
secure
testing
firms
specializing
in
FTIR
measurements,
and
should
increase
the
number
of
available
testing
firms.
New
sources
that
must
meet
the
requirements
of
the
final
rule
are
aware
in
advance
that
their
source
will
be
covered
by
the
final
rule.
The
EPA
feels
that
180
days
is
sufficient
time
to
secure
appropriate
testing
firms
and
to
conduct
the
initial
performance
test
and
feels
that
1
year
to
conduct
the
initial
performance
test
is
not
necessary.
Regarding
the
comment
concerning
seasonal
operations,
new
sources
do
not
have
to
test
until
the
unit
is
operating,
so
seasonal
operation
should
not
be
a
concern
for
new
units.
Also,
for
existing
sources,
EPA
feels
that
seasonal
operation
should
not
be
a
problem
since
the
unit
has
3
years
and
180
days
to
conduct
the
initial
performance
test,
and
surely
the
unit
would
be
operational
within
that
timeframe.
Finally,
the
180
day
time
period
is
consistent
with
the
General
Provisions
of
part
63.

III.
D.
5
Comment:
Two
commenters
(
564
and
565)
requested
that
idle
(
not
in
service)
units
should
have
no
immediate
compliance
obligation
with
the
rule
and
should
be
given
at
least
1
year
from
the
initial
date
of
re­
activation
to
comply
with
the
requirements.

Response:
The
EPA
disagrees
with
the
commenter.
First,
new
stationary
RICE
are
only
required
to
meet
emission
limitations
and
testing
after
startup,
which
is
when
they
are
first
operated.
Existing
engines
that
are
not
in
operation
would
be
considered
limited
use
engines
and
43
are
not
subject
to
immediate
compliance
with
the
emission
limitations.
However,
EPA
feels
that
when
these
engines
are
put
into
more
than
limited
service,
they
should
be
required
to
meet
applicable
standards.
Sources
should
be
well
aware
of
the
possibility
of
their
engines
needing
to
be
put
into
regular
service
and
should
plan
accordingly.

E.
Other
III.
E.
1
Comment:
Two
commenters
(
526,
579)
asked
that
operators
be
allowed
1
year
to
design,
execute
and
submit
results
of
a
testing
program
to
define
operating
parameters.
The
commenters
were
of
the
opinion
that
if
one
should
choose
alternative
formaldehyde
limits
developing
site
specific
operating
conditions
would
be
cumbersome.
The
commenters
stated
that
the
concept
behind
the
petitioning
is
meritorious
in
that
it
provides
flexibility,
many
of
the
requirements
border
on
research­
oriented
items
that
perhaps
only
the
largest
and
most
knowledgeable
operators
are
capable
of
performing.
The
EPA
acknowledges
at
III.
F.
3
of
the
proposed
rule
that
it
wishes
to
encourage
alternative
emission
control
technologies,
hence
the
reason
for
formaldehyde
concentration
limits.

Response:
The
EPA
does
wish
to
encourage
new
and
alternative
emission
control
technology.
However,
the
EPA
anticipates
that
the
information
required
to
be
submitted
with
the
petition
to
the
Administrator
for
approval
of
operating
parameters
would
already
be
available
to
the
owners
and
operators
prior
to
installing
the
new
technology.
In
other
words,
the
owners
and
operators
would
have
some
basis
for
applying
the
new
technology.
Otherwise,
EPA
feels
it
would
not
be
applied.

IV.
EMISSION
LIMITATIONS
A.
Level
IV.
A.
1
Comment:
One
commenter
(
552)
asserted
that
the
emission
limitations
are
too
stringent.
The
commenter
stated
that
the
proposed
emission
standards
were
based
on
information
from
only
five
engines
and
does
not
think
that
the
proposed
percent
reductions
and
emission
standards
reflect
the
actual
performance
possible
from
the
wide
array
of
engine
designs
and
sizes
in
the
marketplace.
For
example,
the
formaldehyde
reduction
standard
for
rich
burn
engines
in
the
proposed
rule
is
set
at
75
percent.
However,
the
data
in
the
docket
show
that
results
from
eight
test
runs
on
two
rich
burn
engines
varied
from
73
to
80
percent.
If
the
reduction
efficiency
on
two
test
engines
under
highly­
controlled
conditions
can
vary
by
such
a
significant
amount
(
and
to
a
level
that
does
not
meet
the
proposed
standard),
then
it
is
highly
likely
that
rich
burn
engines
of
different
size
and
using
different
NSCR
technology
also
would
not
be
able
to
meet
the
standard.
The
EPA
must
consider
the
significant
variability
in
RICE
and
adjust
all
final
emissions
standards
and
reduction
percentages
accordingly.
The
commenter
recommended
that
the
formaldehyde
emission
limits
be
revised
upward
by
10
percent
to
allow
for
variability
in
the
RICE
and
aftertreatment
system
populations.
44
Three
commenters
(
549,
551,
565)
asserted
that
the
MACT
floor
for
existing
4SRB
is
not
representative
of
the
average
emission
limit
achieved
by
the
best
performing
12
percent
of
existing
sources.

One
commenter
(
564)
stated
that
the
emission
standard
for
existing
4SRB
engines
should
be
reassessed
to
be
consistent
with
the
requirements
of
CAA
section
112(
d).
The
commenter
remarked
that
the
Agency
used
the
incorrect
approach
to
set
the
emission
limit
for
existing
4SRB
engines,
which
logically
should
be
lower
percent
removal
than
for
new
4SRB
engines.
It
was
the
commenter's
opinion
that
the
Agency
should
revisit
the
analysis
and
establish
an
emission
limit
for
4SRB
engines
more
consistent
with
the
required
floor­
setting
methodology.

Five
commenters
(
526,
549,
551,
579,
564)
expressed
that
the
same
emission
limitation
for
existing
and
new
4SRB
is
unrealistic.
Commenter
526
recommended
considering
10
percent
less
restrictive
emission
reduction
requirement
for
existing
units.
Another
commenter
(
564)
indicated
that
practically
speaking,
retrofitting
existing
equipment
rarely
achieves
the
optimum
design
available
in
new
equipment.

One
commenter
(
579)
contended
that
350
parts
per
billion
by
volume
dry
basis
(
ppbvd)
is
too
low.
The
chosen
limit
was
achieved
by
the
best
performing
engine
during
CSU
testing
while
for
other
types
of
engines
the
highest
emissions
from
the
performance
range
had
been
chosen
as
the
emissions
limit.

Response:
The
EPA
disagrees
with
comments
that
the
MACT
floor
level
proposed
for
existing
4SRB
engines
is
inconsistent
with
the
statute
or
not
representative
of
the
average
emission
level
achieved
by
the
best
performing
12
percent
of
existing
sources.
The
commenters
do
not
dispute
the
accuracy
of
the
data
used
or
the
representativeness
of
the
engines
tested.
The
commenters
instead
believe
the
manner
in
which
the
EPA
used
the
data
is
not
reflective
of
the
average
of
the
best
performing
12
percent
of
existing
sources.
To
clarify
EPA's
approach
in
the
proposal,
the
EPA
found
the
lowest
percent
reduction
value
for
each
of
the
two
sources
tested,
which
account
for
variability
in
results
for
each
source.
However,
as
EPA
found
that
27
percent
of
the
engines
in
the
subcategory
use
NSCR,
EPA
felt
that
it
was
appropriate
to
use
only
the
higher
of
the
two
values
to
determine
the
MACT
floor
for
existing
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.
This
is
more
closely
analogous
to
the
level
of
the
top
12
percent
of
sources
than
is
a
straight
average
of
the
two
sources.

However,
in
reviewing
EPA's
method
in
response
to
these
comments,
EPA
feels
that
it
would
be
more
appropriate
to
include
in
the
analysis
the
data
from
the
lower
performing
of
the
two
engines
tested,
thus
using
more
than
a
single
data
point
in
determining
the
MACT
floor
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,
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
1
The
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
ppbvd
is
as
follows:
(
highest
tested
ppbvd
of
the
lower
performing
engine)
*
(.
222)
+
(
highest
tested
ppbvd
of
the
higher
performing
engine)
*
(.
778)
=
(
355)(.
222)
+
(
348)(.
778)
=
350.

45
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)
compared
to
the
statutory
level
(
12
percent).
This
leads
to
a
weighted
average
where
the
data
point
for
the
lower
performer
will
be
worth
22
percent
(
50
percent)
(
12/
27)
and
the
level
for
the
higher
performer
will
be
worth
78
percent.

To
be
consistent
with
the
approach
followed
for
other
engine
types,
i.
e.,
establish
emission
limitations
based
on
test
results
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
MACT
floor
of
76
percent
control
efficiency
or
350
ppbvd.
1
Though
the
formaldehyde
reduction
number
differs
slightly
from
the
proposed
level,
it
is
very
close.
The
proposed
level
for
the
alternative
formaldehyde
concentration
emission
limitation
remains
the
same
even
after
following
the
revised
approach.
This
should
not
be
particularly
surprising.
Though
the
emission
values
of
the
two
engines
were
not
identical,
they
were
very
close
and
the
final
value
for
either
engines
generally
rounds
to
the
same
value.

For
new
4SRB
engines,
the
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
EPA
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,
EPA
acknowledges
that
the
proposed
formaldehyde
reduction
should
have
been
set
at
76
percent
reduction
for
4SRB
engines
and
not
75
percent
formaldehyde
reduction
and
has
written
this
in
the
final
rule.

The
commenters
also
seem
to
argue
that
the
MACT
floor
levels
for
existing
engines
must
be
less
stringent
than
those
for
new
engines.
While
the
criteria
for
the
MACT
floor
for
new
engines
is
more
stringent
than
for
existing
engines,
it
is
not
impossible
or
even
illogical
for
the
result
to
be
the
same,
or
at
least
very
close.
In
this
case,
the
best
performing
12
percent
of
engines
use
the
same
control
technology,
and
the
emission
values,
as
well
as
the
emission
reduction
values,
appear
to
be
very
close
for
these
engines.
Therefore,
it
is
not
surprising
that
the
levels
for
the
MACT
floor
for
new
and
existing
engines
should
be
close.
Moreover,
EPA
was
using
a
very
small
data
set
in
setting
the
final
emission
limits,
thus
limiting
the
variation
in
the
data
used.
This
led
to
a
proposed
level
that
used
the
same
calculations
for
determining
the
MACT
floor
for
both
existing
and
new
engines.
The
EPA
has
changed
the
manner
of
calculating
the
MACT
floor
for
existing
engines
for
the
final
rule,
but
the
result
is
still
the
same
as
that
for
new
engines.
Again,
this
is
because
the
results
for
both
engines
were
very
close.
46
Regarding
the
comment
referring
to
the
use
of
the
average
of
the
best
five
performing
sources,
this
is
only
permitted
when
the
category
or
subcategory
has
less
than
30
sources.
This
is
not
the
case
with
this
subcategory.
Given
that
EPA
had
usable
data
from
only
two
sources,
it
is
not
clear
that
averaging
the
two
sources
would
be
appropriate
to
meet
that
requirement.

Regarding
the
comment
that
retrofitting
existing
equipment
rarely
achieved
the
optimum
design
available
in
new
equipment,
the
commenters
provide
no
data
showing
that
emissions
reductions
from
retrofitting
existing
engines
would
be
reduced
compared
to
those
from
new
engines.
Regardless,
the
MACT
floor
for
new
engines
is
not
based
on
the
optimum
possible
design
for
a
new
engine,
but
on
the
best
level
of
control
achieved
in
practice
by
the
best
controlled
similar
source,
whether
retrofitted
or
not.
Similarly,
the
MACT
floor
for
existing
engines
is
based
on
a
specific
formula.
The
EPA
based
the
MACT
floor
for
new
engines
on
the
information
available
to
EPA
from
existing
engines.
While
individual
existing
sources
may
have
some
design
constraints
in
installing
the
emission
control
technology,
there
is
no
evidence
that
the
MACT
floor
is
not
achievable.
The
suggestion
that
is
provided,
a
10
percent
discount
for
existing
units,
without
a
basis
in
the
existing
data,
does
not
appear
consistent
with
the
requirements
of
CAA
section
112(
d).

IV.
A.
2
Comment:
One
commenter
(
527)
indicated
that
there
is
considerable
doubt
about
the
ability
of
an
oxidation
catalyst
to
reduce
the
formaldehyde
concentration
over
long
periods
of
time.
A
technical
paper
presented
at
the
2002
Gas
Machinery
Conference
found
that
the
catalyst
efficiency
for
the
Waukesha
GL
engine
for
formaldehyde
reduces
from
100
percent
to
67
percent
in
only
150
hours
of
operation.

Response:
The
EPA
accounted
for
catalyst
aging
in
setting
the
standard.
In
fact,
the
oxidation
catalysts
used
during
EPA's
testing
at
CSU
were
sufficiently
aged
prior
to
testing.
The
2SLB
engine
catalyst
was
aged
for
236
hours,
the
4SLB
engine
catalyst
was
aged
for
140
hours,
and
the
CI
engine
catalyst
was
aged
for
100
hours.
Industry
representatives
were
in
agreement
that
the
catalysts
were
adequately
aged.
The
industry
testing
EPA
used
in
setting
the
standard
for
4SRB
engines
was
based
on
testing
of
two
4SRB
engines
equipped
with
NSCR.
The
NSCR
catalysts
used
were
appropriately
aged
by
more
than
2
years
prior
to
testing.
Please
refer
to
the
memorandum
entitled
"
Catalyst
Aging
During
EPA
RICE
Testing
at
CSU"
available
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0062
(
A­
95­
35
II­
B­
39)
for
additional
information
regarding
catalyst
aging.

IV.
A.
3
Comment:
One
commenter
(
579)
expressed
support
for
EPA's
attempt
to
stimulate
development
and
eventual
use
of
alternative
control
technologies
with
the
introduction
of
formaldehyde
concentration
limits.
Such
flexibility
has
been
heavily
relied
upon
to
ultimately
meet
the
goals
of
air
quality
management
plans.

Response:
The
commenter's
support
is
acknowledged
by
EPA.

IV.
A.
4
Comment:
One
commenter
(
579)
supports
EPA's
conclusion
that
emission
control
technologies
which
lead
to
reductions
in
formaldehyde
emissions
will
lead
to
reductions
in
47
other
HAP
emissions.

Response:
The
EPA
has
noted
the
commenter's
support.

IV.
A.
5
Comment:
One
commenter
(
565)
expressed
support
for
limiting
the
rule
to
formaldehyde.
The
commenter
also
supported
EPA's
decision
to
use
formaldehyde
as
a
surrogate
for
other
HAP.

Response:
The
EPA
acknowledges
the
commenter's
support.

IV.
A.
6
Comment:
One
commenter
(
565)
urged
EPA
to
interpret
National
Lime
II
as
requiring
it
to
set
emission
standards
only
for
those
HAP
emitted,
or
at
most,
reasonably
expected
to
be
emitted
by
units
in
the
source
category.

Response:
The
standards
promulgated
in
the
final
rule
apply
only
to
pollutants
actually
emitted
by
the
regulated
sources.
This
is
consistent
with
the
comment.

IV.
A.
7
Comment:
One
commenter
(
605)
remarked
that
the
information
in
the
Federal
Register
indicates
that
there
is
not
a
good
correlation
between
CO
reduction
and
formaldehyde
reduction
for
spark
ignited
rich
burn
engines.
The
commenter
was
also
concerned
that
using
CO
reduction
as
a
surrogate
for
formaldehyde
reduction
may
not
be
entirely
appropriate
for
spark
ignited
lean
burn
and
compression
ignited
engines.

Response:
The
commenter
is
correct
in
stating
that
EPA
did
not
determine
that
there
is
a
good
relationship
between
CO
emission
concentration
or
CO
emissions
reductions
and
HAP
emissions
concentrations
or
HAP
emissions
reductions
from
4SRB
engines
equipped
with
NSCR.
The
EPA
therefore
found
it
appropriate
not
to
use
CO
but
to
use
formaldehyde
as
a
surrogate
for
all
HAP
for
4SRB
engines.
However,
for
lean
burn
and
CI
engines,
EPA
was
able
to
determine
that
there
is
in
fact
a
good
relationship
between
CO
emissions
reductions
and
HAP
emissions
reductions
from
2SLB,
4SLB,
and
CI
engines
equipped
with
oxidation
catalyst
controls.
Data
indicate
that
when
CO
emissions
are
reduced,
HAP
emissions
are
also
reduced
in
a
relatively
proportional
manner.
Please
refer
to
the
memorandum
entitled
"
CO
Removal
Efficiency
as
a
Surrogate
for
HAP
Removal
Efficiency
­
Spark
Ignition
Lean
Burn
Clean
Gaseous
Fueled
Engines
and
Compression
Ignition
Engines
Using
an
Oxidation
Catalyst"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35)).
In
addition,
information
received
during
the
public
comment
period
also
supports
this
relationship
and
did
not
change
EPA's
position.
Therefore,
EPA
found
it
appropriate
to
use
CO
emissions
reductions
as
a
surrogate
for
HAP
emissions
reductions
for
lean
burn
and
CI
engines
when
operating
with
an
oxidation
catalyst.

IV.
A.
8
Comment:
One
commenter
(
580)
said
that
the
14
parts
per
million
by
volume
dry
basis
(
ppmvd)
formaldehyde
limit
for
4SLB
engines
is
not
achievable
and
should
be
increased.
The
commenter
stated
that
EPA
based
its
proposed
limit
on
a
small
number
of
tests
on
a
newly
rebuilt
engine
over
a
test
period
of
8.8
hours.
Only
a
single
4SLB
was
tested,
and
it
may
not
be
48
representative
of
engines
of
the
same
type
from
different
manufacturers.
The
period
of
catalyst
aging
was
very
short
compared
to
typical
catalyst
maintenance
intervals,
so
results
may
not
be
representative
of
catalyst
performance
during
normal
catalyst
maintenance
intervals;
and
the
tests
were
performed
within
only
a
single
catalyst
that
may
not
be
representative
of
catalysts
from
different
manufacturers.
Clearly,
all
4SLB
stationary
RICE
cannot
meet
the
emissions
limits
set
by
EPA
in
the
proposed
rule,
particularly
over
normal
catalyst
life
intervals
of
2
to
3
years.
The
EPA
should
incorporate
other
available
test
data
in
the
final
emission
limits
for
4SLB
engines
to
accommodate
the
degradation
in
catalyst
performance
over
the
useful
lifetime
of
the
catalyst.

Response:
The
MACT
floor
for
new
sources
cannot
be
less
stringent
than
the
emission
control
that
is
achieved
in
practice
by
the
best
controlled
similar
source.
The
alternative
formaldehyde
standard
for
4SLB
engines
is
based
on
the
minimum
level
of
control
achieved
by
the
best
controlled
source.
This
approach
takes
into
account
variability
of
the
best
performing
engine.
Furthermore,
EPA
and
industry
representatives
were
in
agreement
that
the
engines
and
catalysts
tested
at
CSU
were
representative
of
engine
and
catalyst
operation
across
the
U.
S.
The
EPA
discussed
catalyst
aging
during
the
EPA
testing
at
CSU
in
response
to
comment
IV.
A.
2.
The
EPA
feels
the
catalyst
was
sufficiently
aged
prior
to
testing
at
CSU.
Industry
representatives
also
agreed
that
the
catalyst
was
adequately
aged.
For
the
reasons
provided,
EPA
feels
that
the
14
ppmvd
formaldehyde
limit
that
was
proposed
for
4SLB
is
appropriate
and
achievable.
The
EPA
recognizes
that
the
alternative
formaldehyde
emission
limitation
is
based
on
a
limited
amount
of
data.
However,
EPA
feels
that
sources
with
a
well
designed
oxidation
catalyst
that
operate
the
equipment
properly
will
be
able
to
meet
the
formaldehyde
concentration.

IV.
A.
9
Comment:
One
commenter
(
588)
agreed
that
an
oxidation
catalyst
represents
the
MACT
floor
for
the
4SLB
engine
category.
For
alternative
controls,
the
commenter
questioned
the
emission
level.
The
commenter
contended
that
the
level
of
control
required
for
alternative
control
options
should
be
similar
to
that
required
by
engines
using
oxidation
catalyst.
Based
on
test
results,
there
are
some
engines
uncontrolled
or
with
minor
formaldehyde
reductions
that
can
meet
the
proposed
level.
Based
on
test
results
and
to
be
consistent
with
the
requirements
set
for
4SRB,
the
formaldehyde
emission
rate
for
4SLB
should
be
set
at
the
lower
level
measured
during
the
CSU
testing
of
6.4
ppmvd.

Response:
The
EPA
disagrees
with
the
commenter.
The
same
methodology
that
was
used
to
set
requirements
for
new
4SRB
engines
were
used
to
set
the
requirements
for
new
4SLB
engines.
The
EPA
utilized
the
same
data
used
to
establish
the
percent
reduction
requirements
to
determine
the
alternative
emission
limitation
requirement.
The
emission
limitation
set
for
new
4SLB
engines
is
based
on
the
minimum
level
of
control
achieved
by
the
best
controlled
4SLB
engine.
This
approach
takes
into
account
the
variability
of
the
best
performing
engine.
The
EPA
therefore
chose
the
highest
controlled
level
and
therefore
also
feels
it
has
appropriately
set
the
alternative
emission
standard
at
14
ppmvd
for
4SLB
engines.

IV.
A.
10
Comment:
One
commenter
603(
D)
said
that
EPA
should
increase
the
specific
formaldehyde
emission
limit
for
new
4SLB
engines
from
14
ppmvd
to
a
more
achievable
rate,
and
suggested
providing
an
alternative
to
meet
a
lower
formaldehyde
limit
with
the
use
of
an
49
oxidation
catalyst.
The
goal
should
be
pollution
prevention,
and
the
standards
should
allow
options
as
to
how
to
achieve
the
desired
result.

Response:
The
EPA
feels
that
based
on
information
available
from
4SLB
testing
at
CSU
that
the
proposed
emission
limit
of
14
ppmvd
for
formaldehyde
is
achievable.
The
commenter
did
not
provide
any
information
to
support
why
they
did
not
believe
the
proposed
limit
is
achievable.
Therefore,
EPA
has
not
changed
the
formaldehyde
limit
of
14
ppmvd
for
4SLB
engines
in
the
final
rule.
The
EPA
is
no
longer
linking
a
particular
control
technology
to
the
standard.
Lean
burn
and
CI
engines
have
the
option
to
either
comply
with
the
CO
percent
reduction
requirement
or
comply
with
the
formaldehyde
concentration
regardless
of
which
control
technique
is
applied.
Similarly,
4SRB
engines
can
comply
with
either
the
formaldehyde
reduction
requirement
or
the
formaldehyde
concentration
requirement
regardless
of
which
control
technique
is
used.

IV.
A.
11
Comment:
One
commenter
(
552)
expressed
that
the
CO
percent
reductions
are
too
stringent.
The
EPA
must
consider
the
significant
variability
in
RICE
and
adjust
the
final
emissions
standards
and
reduction
percentages
accordingly.
Since
there
are
limited
data
available
for
such
an
analysis,
the
commenter
recommended
that,
to
allow
for
variability
in
the
RICE
and
aftertreatment
system
populations,
that
CO
reductions
be
revised
to
50
percent
for
2SLB,
80
percent
for
4SLB,
and
60
percent
for
CI
engines.
The
proposed
relaxation
for
CI
engines
is
particularly
justified
in
light
of
the
very
low
relative
formaldehyde
emissions
from
these
engines.

Response:
The
EPA
disagrees
with
the
commenter.
In
setting
the
CO
percent
reduction
emission
limitation
for
lean
burn
and
CI
engines,
EPA
looked
at
various
scenarios
and
operating
conditions.
These
various
scenarios
and
operating
conditions
represent
the
variability
EPA
expects
would
be
associated
with
engine
and
catalyst
operation
throughout
the
U.
S.
Based
on
testing
results,
EPA
therefore
feels
it
has
appropriately
set
the
CO
percent
reduction
emission
limitation.
As
noted
above,
the
MACT
floor
for
new
engines
must
reflect
the
emission
performance
of
the
best
controlled
similar
engine.
The
EPA
did
adjust
the
CO
percent
reduction
requirement
for
2SLB
from
60
to
58
percent
to
fully
reflect
the
test
results
that
are
the
basis
for
the
standard.
This
is
discussed
further
in
response
to
comment
IV.
D.
3.

IV.
A.
12
Comment:
One
commenter
(
567)
favored
having
a
set
CO
emission
concentration
emission
limit
in
lieu
of
a
CO
emission
reduction
emission
limit.
Similarly,
two
commenters
(
565
and
586)
contended
that
it
should
be
possible
to
set
a
CO
concentration
limit
as
an
alternative
compliance
method
as
was
done
for
formaldehyde.

Response:
The
EPA
was
unable
to
identify
a
relationship
between
CO
concentration
and
HAP
concentration
(
and
therefore
also
formaldehyde
concentration
as
formaldehyde
is
a
surrogate
for
all
HAP)
for
all
subcategories
of
stationary
RICE.
Therefore,
the
EPA
could
not
set
a
CO
concentration
emission
limit.
The
EPA
did
establish
that
there
is
a
good
relationship
between
CO
emissions
reductions
and
HAP
emissions
reductions
from
2SLB,
4SLB,
and
CI
engines
equipped
with
oxidation
catalyst
systems.
As
a
result,
the
EPA
included
a
CO
emission
reduction
emission
limitation
for
these
engines
when
equipped
with
an
oxidation
catalyst.
For
4SRB
engines,
there
is
not
a
good
relationship
between
CO
emissions
reductions
and
HAP
50
emissions
reductions.
For
this
reason,
the
EPA
was
unable
to
use
CO
emission
reduction
as
a
surrogate
for
HAP
emission
reduction
for
4SRB
engines.

IV.
A.
13
Comment:
One
commenter
(
586)
stated
that
at
least
90
percent
CO
control
is
feasible
for
all
types
of
RICE,
as
reflected
in
their
existing
and
proposed
state
of
the
art
manual
for
RICE
which
sets
CO
performance
levels
at
90
percent
reduction
for
all
engine
types.

Response:
The
EPA
disagrees
with
the
commenter
that
90
percent
CO
control
efficiency
is
feasible
for
all
types
of
engines.
Data
show
that
the
performance
level
varies
between
different
subcategories
of
engines.
Therefore,
EPA
has
set
a
different
CO
reduction
emission
limitation
for
each
subcategory.

IV.
A.
14
Comment:
One
commenter
(
565)
stated
that
the
court
has
clearly
decided
that
EPA
can
regulate
the
HAP
emitted
by
the
source
category
by
controlling
the
emission
of
a
surrogate.
In
National
Lime
II,
the
court
upheld
the
use
of
PM
control
as
a
surrogate
for
controlling
HAP,
even
though
PM
is
not
a
HAP,
HAP
was
only
a
small
element
of
the
surrogate
and
the
Agency
did
not
quantify
the
amount
of
control
achieved.
The
commenter
asserted
that
EPA
is
on
firm
ground
controlling
and
measuring
a
surrogate
for
HAP
in
the
source
category,
as
long
as
it
can
show
a
correlation.

Response:
The
EPA
agrees
with
the
commenter.

IV.
A.
15
Comment:
Three
commenters
(
552,
565,
571)
said
that
CO
should
be
a
HAP
surrogate
for
rich
burn
engines
also.
Commenter
571
suggested
that
80
percent
would
be
an
appropriate
efficiency
for
CO.
One
commenter
(
565)
suggested
that
4SRB
engines
should
be
provided
with
a
CO
outlet
concentration
compliance
option
in
addition
to
a
percent
reduction
CO
option.

Response:
The
EPA
determined
that
there
is
not
a
good
relationship
between
CO
emission
concentration
or
CO
emissions
reductions
and
HAP
emissions
concentrations
or
HAP
emissions
reductions
from
4SRB
engines
equipped
with
NSCR.
Therefore
it
is
not
appropriate
to
use
CO
as
surrogate
for
HAP
for
rich
burn
engines.
Since
CO
is
not
an
appropriate
surrogate
for
HAP
for
rich
burn
engines,
4SRB
engines
cannot
be
provided
with
a
CO
outlet
concentration
compliance
option
as
the
CO
concentration
would
not
be
representative
of
HAP
concentration.

IV.
A.
16
Comment:
Several
commenters
(
549,
571,
580,
585,
603(
D))
expressed
that
93
percent
CO
reduction
is
not
achievable.
During
the
public
hearing
a
commenter
stated
that
a
specific
CO
limit
is
more
reasonable.
Two
commenters
suggested
reducing
the
limit
to
require
60
percent
CO
reduction.
One
commenter
recommended
that
the
value
be
set
between
70
and
80
percent
comparable
to
2SLB
and
CI
engines.
Another
commenter
stated
that
EPA
has
not
demonstrated
that
the
catalyst
will
perform
at
this
level
on
a
continuous
basis
considering
fuel
and
lubrication
poisoning.
Finally,
one
commenter
(
585)
said
that
American
Petroleum
Institute/
Gas
Research
Institute
testing
indicated
a
53
to
63
percent
performance.
The
commenter
also
said
that
the
percent
reduction
likely
will
not
be
achievable
with
aged
catalysts.
51
One
commenter
(
549)
had
several
concerns
with
establishing
the
CO
reduction
limit
based
on
the
testing
conducted
at
CSU.
The
concerns
stated
by
the
commenter
include:
only
a
single
engine
for
each
type
was
tested
and
it
may
not
be
representative
of
engines
of
the
same
type
from
different
manufacturers;
the
variables
consisted
only
of
parameters
affecting
HAP
formation
in
the
engine
and
not
necessarily
those
affecting
CO
reduction
across
the
catalyst;
the
engines
were
rebuilt
prior
to
testing
to
represent
new
engines
and
may
not
represent
engine
condition
between
routine
maintenance
intervals;
the
period
of
catalyst
aging
was
very
short
compared
to
typical
catalyst
maintenance
intervals,
hence
results
may
not
be
representative
of
catalyst
performance
during
normal
catalyst
maintenance
intervals;
and
the
tests
were
performed
with
only
a
single
catalyst
that
may
not
be
representative
of
catalysts
from
different
manufacturers.

One
commenter
(
549)
stated
that
catalyst
performance
degrades
over
time
due
to
gas
species
and
concentrations,
thermal
cycling,
chemical
poisoning
and/
or
physical
blocking
caused
by
sulfur,
lubricants,
silica,
etc.
that
enter
the
exhaust
from
the
fuel,
crankcase
and/
or
combustion
air.
Catalyst
life
is
the
dominant
factor
in
the
cost
of
the
control
technology,
since
the
cost
of
replacement
catalyst
modules
is
large
relative
to
other
operating
and
maintenance
costs.
Typically,
oxidation
catalysts
undergo
two
stages
of
deactivation:
a
period
of
rapid
deactivation
as
the
catalyst
adjusts
to
the
thermal
and
gas
conditions,
typically
over
a
period
on
the
order
of
100
hours;
followed
by
a
period
of
slow
deactivation
that
occurs
over
thousands
or
tens
of
thousands
of
hours.
The
duration
of
the
CSU
tests
was
clearly
insufficient
to
address
long­
term
catalyst
deactivation,
and
perhaps
not
even
fully
accounting
for
initial
deactivation.
For
example,
CO
reduction
efficiency
during
the
140
hours
of
catalyst
aging
during
the
4SLB
engine
test
at
CSU
was
still
declining
at
the
end
of
that
period,
suggesting
that
further
deactivation
would
likely
occur
over
time.

Response:
The
EPA
disagrees
with
the
commenter
that
93
percent
reduction
for
CO
is
not
achievable
for
4SLB
engines.
The
93
percent
CO
reduction
emission
limitation
is
based
on
the
minimum
level
of
control
achieved
by
the
4SLB
engine
tested
at
CSU.
The
EPA
chose
the
minimum
efficiency
achieved
as
this
value
takes
into
account
variability
in
performance
of
the
engine
and
engines
operating
across
the
U.
S.,
therefore,
EPA
feels
it
has
appropriately
set
the
emission
limitation
for
4SLB
engines.

As
rationale
for
setting
the
limit
at
60
percent,
the
commenter
cited
a
recent
field
test
of
a
4SLB
engine
where
the
measured
CO
reduction
efficiency
was
53
to
60
percent.
However,
the
commenter
did
not
provide
any
indication
of
what
reduction
efficiency
the
catalyst
was
designed
for,
or
whether
the
catalyst
had
been
properly
maintained
and
cleaned.
The
commenter
also
did
not
identify
the
operating
conditions
under
which
the
test
was
conducted,
for
example
if
the
test
was
conducted
during
high
load
operation.
Moreover,
given
the
results
of
the
CSU
testing,
and
the
standard­
setting
requirements
for
new
engines
under
CAA
section
112(
d),
it
is
not
clear
that
the
results
in
that
test
would
be
relevant
for
standard­
setting
for
new
engines.

Regarding
the
concerns
expressed
by
one
commenter,
EPA
and
industry
representatives
were
in
agreement
that
the
engines
and
catalysts
tested
at
CSU
were
representative
of
engine
and
catalyst
operation
across
the
U.
S.
As
explained
in
the
preamble
to
the
proposed
rule,
the
testing
52
conducted
at
CSU
to
obtain
HAP
and
CO
emissions
data
was
a
joint
EPA­
industry
effort.
Prior
to
testing,
EPA
and
industry
developed
a
list
of
engine
operating
parameters
that
were
known
to
vary
throughout
the
U.
S.
for
each
type
of
engine.
The
engines
and
control
devices
were
tested
at
typical
engine
conditions
in
which
these
operating
parameters
were
varied.
The
variations
in
the
emission
reduction
results
for
each
engine
type
are
due
to
the
variability
of
the
engine
and
control
system
and
include
a
representation
of
the
performance
of
the
best
controlled
source
for
new
engines.
Equipment
manufacturers,
catalyst
vendors,
owners
and
operators,
and
EPA
agreed
that
the
tests
conducted
at
CSU
were
representative
of
typical
engine
operating
conditions
in
the
field
for
varied
engine
and
catalyst
manufacturers.
It
is
believed
that
the
variations
in
the
operating
parameters
affect
both
HAP
formation
and
CO
reduction
across
the
catalyst.
Please
refer
to
the
memorandum
entitled
"
Rationale
for
Development
of
a
Test
Plan
for
RICE"
available
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0048
(
A­
95­
35
II­
B­
26)
for
additional
information
regarding
the
CSU
testing.

The
EPA
disagrees
that
the
catalyst
will
not
perform
at
this
level
on
a
continuous
basis
or
when
it
is
aged.
The
CSU
testing
was
funded
by
several
different
agencies,
and
several
stakeholders
participated
in
the
planning,
preparation
and
execution
of
the
tests.
All
stakeholders
agreed
that
the
catalyst
was
properly
aged
before
testing
was
initiated
on
each
engine.
The
EPA
discussed
catalyst
aging
during
the
testing
at
CSU
in
response
to
a
previous
comment.
The
EPA
feels
the
catalysts
were
sufficiently
aged
prior
to
testing
at
CSU.
It
should
be
noted,
as
discussed
below,
that
sources
may
meet
the
formaldehyde
concentration
standard
to
meet
the
requirements
as
well
as
the
93
percent
CO
reduction
requirement.

In
response
to
the
comment
regarding
long­
term
catalyst
deactivation,
EPA
reemphasizes
that
industry
representatives
that
were
involved
in
the
testing
at
CSU
agreed
that
the
testing
would
be
representative
for
catalyst
performance,
both
short­
term
and
long­
term.
The
EPA
agrees
with
the
commenter
that
there
may
be
two
stages
of
deactivation.
The
first
stage
of
deactivation
may
occur
during
the
first
100
hours,
or
might
occur
as
early
as
after
20
hours
of
operation.
A
second
stage
of
deactivation
may
occur
over
a
period
of
more
than
a
1,000
hours
of
operation.
However,
information
received
from
catalyst
vendors
indicate
that
they
are
able
to
design
the
catalyst
to
achieve
the
guaranteed
percent
reduction
at
the
end
of
the
catalyst
life
(
warranty
period).
The
percent
reduction
may
decline
slightly
in
the
beginning
but
the
catalyst
can
be
designed
to
stabilize
at
the
desired
percent
reduction.
Catalysts
that
can
achieve
emissions
reductions
of
93
percent
or
more
for
the
life
of
the
catalyst
are
within
technological
limits
of
this
technology.
For
these
reasons,
EPA
feels
the
CO
percent
reduction
requirement
of
the
final
rule
is
appropriate
and
justified.

B.
Flexibility/
Options
IV.
B.
1
Comment:
Seven
commenters
(
542,
544,
552,
554,
556,
578,
587)
requested
that
for
lean
burn
engines,
EPA
include
a
percent
reduction
and
concentration
standard
for
both
CO
and
formaldehyde.

Response:
The
EPA
determined
that
there
is
not
a
good
relationship
between
CO
53
emissions
concentration
and
HAP
emissions
concentration
from
lean
burn
engines.
Please
refer
to
the
memorandum
entitled
"
CO
Removal
Efficiency
as
a
Surrogate
for
HAP
Removal
Efficiency
­
Spark
Ignition
Lean
Burn
Clean
Gaseous
Fueled
Engines
and
Compression
Ignition
Engines
Using
an
Oxidation
Catalyst"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35)).
Therefore,
a
CO
concentration
standard
was
not
included
in
the
proposed
rule.
The
commenters
provide
no
evidence
establishing
that
such
a
relationship
exists
in
any
reliable
manner,
especially
for
control
technologies
other
than
oxidation
catalysts.
A
formaldehyde
emission
reduction
alternative
was
not
included
for
lean
burn
engines
as
testing
to
measure
HAP
emissions
is
expensive
and
EPA
knows
of
no
CEMS
which
could
be
used
continuously
to
monitor
all
HAP
emissions.
Therefore
the
EPA
focused
on
CO
emissions
reductions.
Carbon
monoxide
emissions
are
easy
to
measure,
in
addition,
equipment
to
measure
CO
emissions
is
readily
available.
The
EPA
found
that
there
is
a
good
relationship
between
CO
emissions
reductions
and
HAP
emissions
reductions
from
2SLB,
4SLB,
and
CI
stationary
engines
using
oxidation
catalyst
control.
Therefore
it
was
appropriate
to
include
a
CO
percent
reduction
emission
limitation
for
these
engines
when
using
oxidation
catalyst
emission
control
in
the
rule.

IV.
B.
2
Comment:
Ten
commenters
(
526,
542,
544,
554,
555,
556,
578,
579,
587,
583)
asked
that
EPA
allow
sources
to
choose
either
percent
reduction
or
final
concentration
to
comply
with
irrespective
of
the
control
technique
employed.

Response:
The
EPA
agrees
with
the
commenters,
and
feels
it
is
appropriate
to
allow
sources
to
choose
either
the
percent
reduction
or
formaldehyde
concentration
outlet
limit
to
demonstrate
compliance
irrespective
of
the
control
technique
employed.
The
EPA
has
specified
this
flexibility
in
the
final
rule.

IV.
B.
3
Comment:
One
commenter
(
581)
requested
that
EPA
delete
the
operating
limitation
requirements
and
modify
the
emissions
limitations.
The
commenter
also
suggested
revising
Table
2A
of
the
proposed
rule
to
offer
flexibility.

Response:
The
commenter
did
not
provide
any
rationale
supporting
why
EPA
should
delete
the
operating
limitation
requirements
and
modify
the
emissions
limitations.
In
addition,
the
commenter
did
not
specifically
indicate
to
what
degree
the
emissions
limitations
should
be
modified.
Lastly,
the
commenter
did
not
provide
enough
information
regarding
suggested
revisions
to
Table
2A
of
the
proposed
rule
to
offer
flexibility.
Therefore,
the
EPA
cannot
address
this
comment.

IV.
B.
4
Comment:
One
commenter
(
565)
urged
EPA
to
promulgate
an
equipment
and
work
practice
standard
to
ensure
good
combustion
as
an
alternative
to
the
CO
and
formaldehyde
emissions
limitation.
The
commenter
suggested
the
following:

(
1)
The
installation
of
a
properly
designed
catalyst
system
commensurate
to
the
type
and
size
of
the
engine
and
meeting
the
design
specifications
established
by
EPA
in
the
rule.
This
would
be
demonstrated
by
statements
from
the
manufacturer.
54
(
2)
Operation
of
the
engine
and
catalyst
system
within
the
design
specifications
of
the
catalyst.

(
3)
An
initial
performance
test
to
verify
that
the
catalyst
system
operation
meets
the
design
specifications
established
by
EPA
in
the
rule.

(
4)
Replacement
of
the
catalysts
following
vendor
recommendations.

(
5)
Initial
compliance
would
be
demonstrated
by
an
owner
or
operator
certification
that
the
design
of
the
catalyst
system
meets
the
design
specifications
established
by
EPA
in
the
rule
and
the
results
of
the
initial
performance
test.

(
6)
Continuous
compliance
would
be
demonstrated
by
the
owner
and
operator
maintaining
records
and
certifying
that
the
catalyst
system
was
maintained
on
the
engine
during
the
compliance
period,
that
the
catalyst
was
still
within
the
vendor's
recommendation
or
that
the
catalyst
was
changed
within
the
vendor's
recommendation,
and
that
any
catalyst
inlet
temperature,
engine
load
or
fuel
flow
alarm
issues
were
responded
to
and
addressed
in
a
timely
manner
according
to
a
written
SSM
plan.

Another
commenter
(
569)
also
requested
that
emission
limitations
for
CO
emissions
and
formaldehyde
should
be
replaced
with
equipment
and
work
practice
requirements.
Commenter
569
recommended
similar
equipment
and
work
practice
requirements
as
commenter
565.

Response:
The
CAA
section
112(
h)
does
not
allow
EPA
to
promulgate
a
design,
equipment,
work
practice,
or
operational
standard
in
lieu
of
an
emission
standard
unless
EPA
finds
that
it
is
not
feasible
to
prescribe
or
enforce
an
emission
standard
for
control
of
the
HAP
in
question,
and
CAA
section
112(
h)(
2)
has
very
specific
criteria
that
EPA
must
meet
to
make
that
showing.
The
commenter
does
not
argue
that
such
a
standard
is
infeasible
for
controlling
HAP
from
this
source
category
and
EPA
does
not
believe
the
criteria
of
CAA
section
112(
h)(
2)
can
be
met.
Therefore,
work
practice
standards
would
not
be
appropriate
for
the
final
rule.

C.
Compression
Ignition/
DPF
IV.
C.
1
Comment:
Two
commenters
(
526,
546)
argued
that
the
proposed
rule
does
not
recognize
DPF
as
a
significantly
more
effective
control
device
for
reducing
diesel
exhaust
emissions
compared
to
diesel
oxidation
catalysts
(
DOC).
One
commenter
(
582)
asked
that
the
final
rule
require
the
use
of
particulate
traps
on
diesel
engines.
Another
commenter
(
555)
expressed
concern
with
the
interaction
of
control
equipment
with
diesel
particulate
traps.
One
commenter
(
586)
indicated
that
DPF
can
reduce
diesel
PM
by
at
least
80
percent.
According
to
the
commenter,
these
traps
can
reduce
CO
by
at
least
90
percent.

Response:
The
commenters
indicate
that
DPF
are
effective
at
reducing
diesel
exhaust
emissions
or
diesel
particulates.
These
are
not
HAP
listed
pursuant
to
section
112(
b)
of
the
CAA
55
and,
therefore,
are
not
the
pollutants
that
the
final
rule
is
targeting
specifically.
The
EPA
has
recently
received
a
request
to
list
diesel
exhaust
pursuant
to
section
112(
b)
of
the
CAA
and
is
currently
reviewing
that
request.

At
the
time
of
proposal
EPA
investigated
DPF.
However,
at
the
time
of
this
investigation,
the
effectiveness
of
DPF
on
listed
HAP
emissions
from
stationary
sources
had
not
been
demonstrated,
and
the
technology
had
only
been
applied
to
a
handful
of
stationary
RICE.
They,
therefore,
were
not
appropriate
as
a
MACT
floor
technology.
The
EPA
examined
DPF
for
their
ability
to
reduce
listed
HAP
and
their
cost
effectiveness.
The
EPA
concluded
that
there
were
no
data
to
show
that
this
technology
would
be
more
effective
at
reducing
listed
HAP
than
oxidation
catalysts.
The
EPA
also
noted
that
this
technology
was
more
expensive
than
oxidation
catalysts,
so
EPA
did
not
use
this
technology
as
a
basis
for
the
proposed
MACT
levels.
However,
the
proposal
did
allow
the
use
of
technologies
other
than
oxidation
catalysts,
including
DPF,
to
meet
the
MACT
requirements,
which
are
generally
numerical,
though
there
were
certain
compliance
options
that
differed
depending
on
the
emission
control
used
on
the
engine.
Since
proposal,
EPA
has
received
new
information
regarding
DPF
resulting
in
EPA
reevaluating
the
feasibility
of
applying
DPF
to
stationary
RICE.
This
information
can
be
accessed
through
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0546
(
A­
95­
35
IV­
D­
44).
In
addition,
the
final
rule
eliminates
all
provisions
linking
the
standard
to
any
particular
control
technology.
Sources
are
free
to
choose
any
compliance
option
irrespective
of
the
control
technique
applied.
The
EPA
has
no
reason
to
believe
that
DPF
are
incompatible
with
oxidation
catalysts
or
that
they
cannot
be
used
instead
of
oxidation
catalysts.
In
the
context
of
its
mobile
source
regulations,
EPA
has
found
that
DPF
can
be
incorporated
with
other
emission
control
devices
without
compatibility
problems.
The
EPA
agrees
with
the
commenter
that
DPF
may
be
able
to
reduce
PM
by
at
least
80
percent
and
that
they
might
be
able
to
also
reduce
CO
by
at
least
90
percent,
at
least
in
certain
instances,
though
EPA
has
determined
that
these
reductions
can
only
be
reliably
achieved
using
ultra
low
sulfur
fuel
(
15
ppm
sulfur
content
by
weight).
However,
the
EPA
does
not
have
any
actual
test
data
showing
that
DPF
can
reliably
reduce
HAP
emissions
from
stationary
CI
engines
at
a
level
beyond
that
already
required
by
the
final
rule.
In
particular,
EPA
does
not
have
data
regarding
actual
use
of
these
devices
on
stationary
RICE,
or
under
the
range
of
operating
parameters
reasonably
expected
for
such
engines.
Also,
the
ultra
low
sulfur
fuel
(
15
ppm
sulfur
content
by
weight)
needed
for
this
technology
is
not
yet
available
in
sufficient
quantities
in
the
U.
S.
The
EPA
therefore
has
determined
that
there
is
currently
not
enough
information
regarding
DPF
as
applied
to
HAP
emissions
from
stationary
CI
engines
on
which
to
base
the
standard
for
the
final
rule.

IV.
C.
2
Comment:
One
commenter
(
582)
urged
EPA
to
rationalize
its
policy
and
address
the
serious
public
health
impacts
associated
with
diesel­
powered
RICE
by
establishing
rigorous
PM
and
clean
fuel
requirements
in
the
final
rule.

Response:
The
EPA
appreciates
the
comments
regarding
pollution
from
diesel­
powered
stationary
RICE.
While
EPA
agrees
that
diesel
engines
emit
pollutants
of
concern
beyond
those
covered
in
the
final
rule,
EPA
does
not
feel
it
would
be
appropriate
to
establish
diesel
PM
or
clean
fuel
requirements
in
the
rule.
The
final
rule
is
a
relatively
narrow
rule,
regulating
only
listed
HAP
56
from
stationary
RICE.
As
noted
in
the
response
to
comment
X.
B.
4,
diesel
PM
is
not
currently
listed
as
a
HAP
under
section
112
of
the
CAA.
While
regulation
of
diesel
PM
may
be
appropriate
in
the
long­
term,
either
as
a
criteria
pollutant
or
as
a
listed
HAP,
the
EPA
does
not
feel
that
the
final
rule,
which
proposed
only
to
regulate
HAP
already
listed
under
CAA
section
112,
is
the
appropriate
place
to
promulgate
final
regulations
affecting
criteria
pollutants
and
precursors
(
like
PM
or
NOx).
Similarly,
the
commenter
does
not
provide
an
explanation
of
the
need
to
regulate
diesel
fuel,
except
as
it
affects
PM
emissions.
Therefore,
EPA
is
not
taking
any
final
action
with
regard
to
these
issues
in
the
final
rule.

IV.
C.
3
Comment:
One
commenter
(
605)
noted
that
using
an
oxidation
catalyst
for
a
CI
engine
is
not
likely
to
produce
the
continuous
HAP
reductions
if
nothing
is
done
to
treat
the
particulate
emissions
from
the
diesel
fuel
to
keep
the
catalyst
from
fouling.
In
order
for
the
oxidation
catalyst
to
work
properly,
either
very
low
sulfur
diesel
and/
or
an
add­
on
DPF
should
be
used.
This
control
technique
would
also
address
the
issue
of
HAP
in
diesel
particulate
emissions.

Response:
The
EPA
does
not
necessarily
disagree
with
the
commenter
but
EPA
expects
owners
and
operators
to
use
low
sulfur
fuel.
The
EPA
is
aware
of
oxidation
catalysts
that
are
compatible
with
higher
sulfur
fuel,
e.
g.,
3000
ppm
sulfur
fuel.
Sulfur
tolerant
catalysts
are
on
the
market
and
do
not
need
low
sulfur
fuel
to
operate.
Low
sulfur
fuel
is
readily
available
in
the
continental
U.
S.
The
EPA
expects
there
may
be
a
few
remote
areas
of
Alaska
where
low
sulfur
fuels
may
not
be
available;
however,
EPA
expects
that
these
areas
would
not
be
located
a
major
sources
of
HAP
and
therefore
not
covered
by
the
final
rule.

D.
Other
IV.
D.
1
Comment:
Two
commenters
(
549,
580)
contended
that
routing
engine
exhaust
to
the
flame
zone
of
a
furnace,
process
heater
or
boiler
should
be
considered
adequate
control
of
organic
HAP.

Response:
The
EPA
does
not
agree
that
routing
the
engine
exhaust
stream
to
a
boiler
or
process
heater
should
be
considered
an
acceptable
control
technology
based
on
the
record.
The
commenters
have
not
provided
any
supporting
information
to
indicate
that
routing
the
engine
exhaust
stream
to
a
boiler,
process
heater,
or
other
waste
heat
recovery
unit
would
control
HAP
emissions
in
this
application.
Moreover,
it
is
unclear
how
compliance
with
these
MACT
requirements
could
be
established
using
this
approach.

IV.
D.
2
Comment:
Two
commenters
(
552,
555)
stated
that
some
allowance
for
catalyst
aging
should
be
considered
when
setting
the
standard.
Commenter
552
stated
that
emission
limits
and
reductions
are
not
reasonable
for
aged
catalysts.

Response:
The
EPA
did
consider
catalyst
aging
in
setting
the
standard.
The
catalysts
used
during
EPA's
testing
of
three
engines
(
2SLB,
4SLB,
and
CI)
were
burned
in
before
each
test.
This
ensured
that
the
catalyst's
efficiency
represented
the
efficiency
achieved
by
mature
catalysts
installed
on
engines
in
industry.
Representatives
from
industry
agreed
that
the
catalysts
57
were
sufficiently
aged
prior
to
the
beginning
of
testing.
Therefore,
EPA
feels
that
the
emission
limits
and
reductions
are
reasonable
for
aged
catalysts
and
that
the
limits
that
are
set
are
reasonable.
Please
refer
to
the
memorandum
entitled
"
Catalyst
Aging
during
EPA
RICE
Testing
at
CSU"
available
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0062
(
A­
95­
35
II­
B­
39)
for
additional
information
regarding
catalyst
aging
at
CSU.
The
industry
testing
EPA
used
in
setting
the
standard
for
4SRB
engines
was
based
on
testing
of
two
4SRB
engines
equipped
with
NSCR.
The
NSCR
catalysts
used
were
appropriately
aged
by
more
than
2
years
prior
to
testing.

IV.
D.
3
Comment:
Five
commenters
(
542,
554,
556,
578,
587)
sought
adjustment
of
the
MACT
emission
limitations
to
fully
reflect
the
test
results
that
are
the
basis
for
the
standard.
One
commenter
(
587)
indicated
that
the
CO
percent
reduction
standard
for
2SLB
engines
should
be
adjusted
to
58
percent
to
reflect
the
lowest
percent
reduction
achieved
during
the
EPA­
sponsored
emission
testing
at
the
CSU
Engine
Lab,
which
is
the
basis
for
the
2SLB
standards.
The
formaldehyde
percent
reduction
standard
for
4SRB
engines
should
be
adjusted
to
73
percent
to
reflect
the
lowest
percent
reduction
achieved
during
the
industry­
sponsored
testing,
which
is
the
basis
for
the
4SRB
emission
standards.
Similarly,
the
formaldehyde
concentration
standard
for
4SRB
engines
should
be
adjusted
to
370
ppbvd
at
15
percent
oxygen
to
reflect
the
highest
post­
NSCR
concentration
of
formaldehyde.

Response:
The
EPA
agrees
with
the
commenter
that
the
CO
percent
reduction
standard
for
2SLB
should
be
adjusted
to
58
percent
to
fully
reflect
the
possible
variation
for
the
best
performing
sources
for
these
engines.
The
EPA
has
made
this
adjustment
in
the
final
rule
to
fully
reflect
the
test
results
obtained
for
the
2SLB
engine
tested
at
CSU.
The
EPA
proposed
an
alternative
formaldehyde
emission
limitation
of
17
ppmvd
for
new
2SLB
engines
in
the
proposal.
The
concentration
for
the
formaldehyde
emission
limitation
was
based
on
the
minimum
level
of
control
achieved
by
the
best
controlled
source.
This
approach
takes
into
account
the
variability
of
the
best
performing
engine.
The
formaldehyde
emissions
at
CSU
ranged
from
7.5
ppmvd
to
17
ppmvd.
Therefore,
EPA
chose
17
ppmvd
at
proposal.
The
17
ppmvd
formaldehyde
concentration
was
based
on
a
run
conducted
at
low
load
(
69
percent).
After
reviewing
EPA's
approach
at
proposal,
EPA
has
found
it
inconsistent
to
establish
the
alternative
formaldehyde
emission
limitation
based
on
the
level
achieved
during
a
low
load
test.
The
approach
that
EPA
has
used
for
other
engine
types
in
establishing
the
alternative
emission
limitations
was
to
establish
the
limits
based
on
high
loads
and
to
require
compliance
at
high
loads.
The
expected
trend
is
for
the
HAP
concentration
to
generally
increase
with
decreasing
load;
however,
EPA
does
not
have
sufficient
data
to
take
the
effect
of
load
into
account
in
establishing
the
alternative
emission
limitations.
Because
of
this,
the
emission
limitations
are
based
on
performance
at
high
loads.
The
EPA
expects
that
if
the
emission
limitations
are
achieved
at
high
load,
then
the
technology
will
be
operating
appropriately
and
will
also
operate
appropriately
at
lower
loads.
To
be
consistent,
the
EPA
has
established,
in
the
final
rule
an
alternative
formaldehyde
emission
limit
for
new
2SLB
engines
of
12
ppmvd.
This
number
is
based
on
the
minimum
level
of
control
achieved
by
the
best
performing
engine
at
high
load
conditions.
The
EPA
has
specified
in
the
final
rule
that
performance
tests
must
be
conducted
at
high
load
conditions,
defined
as
100
percent
±
10
percent.
If
a
source
has
demonstrated
compliance
with
the
emission
limit
at
high
loads
it
is
assumed
that
58
the
technology
is
operating
appropriately
and
will
also
operate
appropriately
at
lower
loads.
Sources
are
not
required
to
meet
the
emission
limitation
at
low
load.

As
described
in
the
preamble
to
the
proposed
rule,
EPA
reviewed
emissions
data
from
an
industry
sponsored
formaldehyde
emission
test
conducted
on
two
4SRB
engines.
The
EPA
selected
the
best
performing
engine
based
on
the
highest
average
formaldehyde
percent
reduction.
The
average
reduction
was
79
percent
for
that
engine;
however,
to
establish
variability
EPA
looked
at
each
of
the
12
individual
test
runs
performance
on
that
engine.
The
percent
reduction
varied
from
75
percent
to
81
percent.
At
proposal,
EPA
selected
75
percent
for
the
MACT
floor.
However,
since
proposal,
EPA
has
reviewed
the
method
it
used
to
set
the
MACT
floor
for
existing
4SRB
engines.
The
EPA
feels
it
would
be
more
appropriate
to
include
in
the
analysis
the
data
from
the
lower
performing
engine,
thus
using
more
than
a
single
data
point
in
determining
the
MACT
floor
for
existing
4SRB
engines.
The
revised
approach
was
discussed
in
detail
in
response
to
comment
IV.
A.
1.
In
the
response
to
comment
IV.
A.
1,
EPA
described
its
revised
approach
which
takes
into
account
the
performance
of
both
engines
tested,
using
a
weighted
average
where
the
data
point
for
the
lower
performer
will
be
worth
22
percent
and
the
level
for
the
higher
performer
will
be
worth
78
percent.
In
addition,
to
be
consistent
with
the
approach
followed
for
other
engine
types,
EPA
has
excluded
runs
conducted
at
low
loads
in
setting
the
MACT
floor.
As
previously
indicated
elsewhere
in
this
document,
since
the
MACT
floor
is
based
on
emissions
data
from
runs
at
high
loads,
performance
tests
must
be
conducted
at
high
load
conditions,
defined
as
100
percent
load,
±
10
percent.
The
commenter
stated
that
the
formaldehyde
percent
reduction
standard
for
existing
4SRB
engines
should
be
adjusted
to
73
percent
to
reflect
the
lowest
percent
reduction
achieved
during
the
industry­
sponsored
testing.
Although
the
commenter
is
correct
in
stating
that
73
percent
formaldehyde
reduction
was
the
lowest
average
reduction,
73
percent
reduction
was
achieved
during
a
run
that
was
not
conducted
at
high
load.
For
this
reason,
it
is
not
appropriate
to
use
the
73
percent
formaldehyde
reduction
in
the
MACT
floor
analysis.
Similarly,
the
run
where
the
formaldehyde
concentration
was
measured
at
370
ppbvd
was
also
not
conducted
at
high
load,
and
was,
therefore,
not
used
in
EPA's
analysis
of
the
MACT
floor
for
existing
4SRB
engines.

IV.
D.
4
Comment:
Two
commenters
(
549,
559)
requested
that
the
burn­
in
period
during
commissioning
of
new
or
rebuilt
engines
should
be
exempted
from
emission
limits.
Catalyst
manufacturer
warrantees
typically
require
a
burn­
in
period
for
new
and
rebuilt
engines
prior
to
placing
the
catalyst
on
stream.
This
is
intended
to
allow
seating
of
critical
engine
components
(
e.
g.,
piston
rings).
Catalyst
placed
on
stream
before
this
burn­
in
period
is
subject
to
physical
damage
from
engine
backfire
and
poisoning
and
or
fouling
from
crankcase
oil
blow­
by.
The
EPA
has
acknowledged
this
need
in
a
prevention
of
significant
deterioration
and
title
V
permit
by
including
the
following
language:

"
The
permittee
shall
be
allowed
to
operate
the
replacement/
overhauled
engine
without
the
use
of
the
catalytic
converter
assembly
for
a
period
not
to
exceed
200
hours
from
the
engine
startup,
unless
a
longer
time
period
has
been
approved
by
EPA,
in
writing."

The
commenters
recommended
that
deviating
from
the
emissions
limits
during
the
burn­
in
period
59
or
the
first
200
hours
of
operation
of
a
new
or
rebuilt
RICE
not
be
considered
a
violation.
The
commenters
recommended
that
a
statement
be
added
at
§
63.6640(
d)
that
deviating
from
the
emissions
limits
during
the
burn­
in
period
or
the
first
200
hours
of
operation
of
a
new
or
rebuilt
RICE
is
not
a
violation.

Response:
The
EPA
agrees
with
the
commenter
that
an
engine
burn­
in
period
of
200
hours
is
appropriate
prior
to
installing
the
catalyst
to
prevent
damage
to
the
catalyst.
The
EPA
has,
therefore,
specified
that
new
or
rebuilt
engines
may
operate
for
up
to
200
hours
prior
to
installing
the
catalyst
in
the
final
rule
and
that
this
will
not
be
considered
a
violation.
However,
sources
have
180
days
after
the
compliance
date
specified
for
their
source
to
conduct
the
performance
test
and
initial
compliance
demonstration
and
the
200
hours
of
burn­
in
time
must
be
conducted
within
the
these
180
days.

IV.
D.
5
Comment:
One
commenter
(
588)
did
not
agree
with
EPA's
determination
of
the
MACT
floor
for
4SLB
RICE.
The
database
used
to
determine
the
MACT
floor
is
based
on
pre­
1999
information
and
includes
542
engines
from
Wyoming.
Since
1999,
Wyoming
has
permitted
2,100
4SLB
engines.
Approximately
62
percent
of
the
greater
than
500
HP
4SLB
permitted
since
1999
have
been
required
to
be
equipped
with
oxidation
catalyst
to
control
formaldehyde.
The
EPA
reports
the
number
of
existing
4SLB
used
in
determining
the
MACT
floor
at
4,149.
Including
the
4SLB
engines
greater
than
500
HP
permitted
since
1999
in
Wyoming,
the
total
is
5,664.
Of
this
total,
935
engines
have
permit
conditions
requiring
oxidation
catalyst
to
control
formaldehyde,
which
is
16.5
percent
of
the
total.
Section
112(
d)
of
the
CAA
requires
the
emission
standard
for
existing
sources
be
no
less
stringent
than
the
emission
limitation
achieved
by
the
best
performing
12
percent
of
existing
sources.
The
commenter
contended
that
the
database
used
to
determine
the
MACT
floor
is
incomplete,
and
EPA
must
reevaluate
the
MACT
floor
including
permitting
actions
post
1998.

Response:
The
EPA
contacted
the
commenter
who
submitted
this
comment.
The
commenter
stated
that
mostly
all
of
the
engines
that
have
been
permitted
are
minor
sources
of
HAP.
During
submittal
of
the
comment,
the
commenter
also
provided
EPA
with
a
database
of
permitted
4SLB
engines.
The
EPA
reviewed
the
database
submitted
by
the
commenter
and
calculated
facility
wide
formaldehyde
emissions
using
worst
case
assumptions
of
operating
8,760
hrs/
yr
for
all
engines
at
the
site.
Calculations
showed
that
only
one
facility
exceeded
the
major
source
threshold
for
a
single
HAP
of
10
tpy.
Since
the
4SLB
engines
permitted
in
Wyoming
are
nearly
all
at
minor
sources
of
HAP,
it
is
not
accurate
to
add
these
sources
to
the
determination
of
the
average
of
the
best
performing
12
percent
of
existing
sources
from
the
source
category.
The
determination
of
the
average
of
the
best
performing
12
percent
of
existing
sources
must
be
based
on
the
sources
regulated.
Since
the
final
rule
only
covers
major
sources,
it
is
not
appropriate
to
include
the
minor
source
engines
permitted
to
require
oxidation
catalyst
in
Wyoming.
Moreover,
the
calculation
of
the
MACT
floor
does
not
require
that
EPA
include
reductions
that
were
implemented
within
18
months
of
the
proposal,
or
30
months
of
the
final
rule.
It
is
not
clear
how
many
of
the
engines
the
commenter
discusses
were
equipped
with
oxidation
catalysts
during
that
period.
Therefore,
the
EPA
has
not
reevaluated
the
floor
for
existing
4SLB
engines.
The
MACT
floor
of
existing
4SLB
engines
remains
at
no
emissions
reductions.
60
IV.
D.
6
Comment:
One
commenter
(
528)
stated
that
emission
standards
should
be
cut
by
50
percent.

Response:
The
EPA
is
unable
to
address
this
comment
as
the
commenter
did
not
provide
any
information
supporting
the
request
to
cut
emissions
standards
by
50
percent.

IV.
D.
7
Comment:
One
commenter
(
588)
responded
to
EPA's
request
for
comment
on
the
percentage
of
4SRB
engines
that
would
choose
to
install
NSCR
HAP
controls
rather
than
other
HAP
controls.
For
new
engines,
the
commenter
stated
that
100
percent
of
the
4SRB
greater
than
500
HP
would
use
NSCR
to
meet
Best
Achievable
Control
Technology
(
BACT)
for
NOx.
The
commenter
did
not
have
a
feel
for
existing
engines,
but
there
is
a
high
possibility
that
an
oxidation
catalyst
would
be
installed
due
to
cost.
An
air/
fuel
ratio
controller
would
have
to
be
installed
with
NSCR.
An
oxidation
catalyst
can
be
operated
with
an
air/
fuel
ratio
controller.

Response:
The
commenter's
response
to
EPA's
request
for
comment
on
this
issue
is
appreciated.

IV.
D.
8
Comment:
One
commenter
(
553)
expressed
support
for
EPA's
decision
not
to
set
"
beyond­
the­
floor"
emissions
limitations
for
certain
categories
and
subcategories
of
RICE.
The
EPA's
declination
to
set
beyond­
the­
floor
emission
standards
for
certain
new
and
existing
RICE
is
a
valid
exercise
of
the
Agency's
discretion.
The
commenter
said
that
EPA
has
properly
conducted
cost­
benefit
analysis
where
it
had
sufficient
data
to
do
so
and
determined
that
more
stringent
standards
were
inefficient;
the
Agency
has
properly
declined
to
set
more
stringent
standards
where
it
lacks
sufficient
data
to
assess
the
attendant
costs
and
benefits;
and
EPA
has
properly
declined
to
require
fuel­
switching
where
it
has
determined
that
doing
so
would
defeat
EPA's
purpose
in
creating
the
subcategory
and
would
be
environmentally
unsound.

Response:
The
EPA
agrees
in
general
with
the
comment
that
EPA's
decision
not
to
set
"
beyond­
the­
floor"
emissions
limitations
in
this
instance
is
appropriate.

IV.
D.
9
Comment:
One
commenter
(
565)
asserted
that
EPA
retains
the
discretion
to
devise
a
"
floor"
methodology,
as
long
as
it
provides
a
reasonable
estimate
of
what
the
best
performers
achieve.
The
EPA
still
retains
the
authority
to
discover
what
the
best
performers
achieve
and
HOW
they
achieve
it
as
part
of
their
floor
analysis.
If
that
methodology
is
reproducible
by
others,
then
it
is
a
legitimate
basis
for
the
MACT
floor
standards.

Response:
The
EPA
feels
the
MACT
floors
are
based
on
evidence
indicating
that
the
MACT
technology
significantly
controls
emissions
and
that
this
approach
is
a
reasonable
one
for
determining
the
performance
of
the
best
performing
units.
The
EPA
feels,
and
the
analysis
indicates,
that
the
levels
required
by
the
MACT
standards
are
reproducible
by
others.

IV.
D.
10
Comment:
One
commenter
(
565)
stated
that
EPA
can
establish
a
"
no
control"
limitation.
This
is
because
the
statute
requires
EPA
to
set
standards
that
are
duplicable
by
others.
In
National
Lime
II,
the
court
threw
out
EPA's
determination
of
a
no
control
floor
because
it
was
61
based
on
a
control
technology
approach.
And
in
CKRC,
the
court
stated
that
EPA
must
look
at
what
the
best
performers
achieve,
regardless
of
how
they
achieve
it.

One
commenter
(
553)
asserted
that
EPA
has
ample
legal
authority
to
set
the
MACT
floor
at
"
no
emissions
reductions"
for
certain
engines.
The
commenter
provided
the
following
basis
for
its
comment:

(
1)
Section
112(
d)
of
the
CAA
requires
EPA
to
set
emissions
standards
in
terms
of
emissions
reductions.

(
2)
EPA's
determination
that
the
MACT
floor
for
certain
RICE
is
"
no
emissions
reduction"
is
lawful
because
EPA
determined
that
the
best­
performing
sources
were
not
achieving
emissions
reduction
through
the
use
of
an
emission
control
system
and
there
were
no
other
methods
by
which
RICE
could
reduce
HAP
emissions.

(
3)
EPA's
MACT
Floor
methodology
is
proper
under
the
D.
C.
Circuit
Holdings
in
Cement
Kiln
Recycling
and
National
Lime
Association.

(
4)
Setting
emissions
standards
on
the
basis
of
actual
emissions
data
alone
where
facilities
have
no
way
of
controlling
their
HAP
emissions
would
contravene
the
plain
statutory
language
as
well
as
Congressional
intent
that
affected
sources
no
be
forced
to
shut
down.

Response:
The
EPA
agrees
in
general
with
the
comment
that
EPA
can
establish
a
MACT
floor
based
on
no
emission
reduction.

V.
MONITORING,
RECORDKEEPING,
REPORTING
A.
CO
CEMS
V.
A.
1
Comment:
Twelve
commenters
(
542,
548,
551,
552,
553,
554,
583,
556,
570,
578,
587,
603(
D))
contended
that
the
CO
CEMS
requirement
for
large
lean
burn
engines
is
unreasonable.
The
commenters
stated
that
parameter
monitoring
and
periodic
testing
should
be
offered
to
CO
monitoring
on
all
lean
burn
engines.
One
commenter
(
552)
noted
that
given
that
the
best
available
emissions
control
technology
for
RICE
is
a
passive
catalyst
system
and
that
the
operator
cannot
reduce
or
improve
HAP
removal
efficiency,
simplified
and
less
costly
environmental
monitoring
requirements
should
be
adopted.

Response:
The
EPA
now
feels
that
the
proposed
requirement
for
2SLB,
4SLB,
and
CI
engines
5,000
HP
or
above
complying
with
the
requirement
to
reduce
CO
emissions
using
an
oxidation
catalyst
to
use
CO
CEMS
is
unnecessary
and
inappropriate.
The
costs
associated
with
a
CO
CEMS
is
estimated
by
EPA
to
be
over
$
200,000
in
capital
costs
and
nearly
$
60,000
in
62
annual
costs.
The
EPA
considers
these
costs
to
be
excessive.
For
these
reasons,
EPA
feels
it
is
not
appropriate
to
include
a
requirement
for
large
lean
burn
and
large
CI
engines
to
install
CO
CEMS
in
the
final
rule.
The
EPA
feels
that
the
combination
of
periodic
stack
testing
and
parameter
monitoring
is
a
proper
and
reasonable
alternative
for
large
engines.
The
semiannual
testing
of
CO
will
ensure,
on
an
ongoing
basis,
that
the
source
is
meeting
the
CO
percent
reduction
requirement.
In
addition
to
semiannual
stack
testing,
2SLB,
4SLB,
and
CI
engines
meeting
the
CO
percent
reduction
requirement
and
using
an
oxidation
catalyst
must
continuously
monitor
and
maintain
the
catalyst
inlet
temperature
as
well
as
maintain
and
monitor
the
pressure
drop
across
the
catalyst
monthly.
These
parameters
serve
as
surrogates
of
the
oxidation
catalyst
performance
and
by
monitoring
and
maintaining
these
parameters,
continuous
compliance
between
semiannual
stack
testing
will
be
ensured.
The
EPA
is
including
CO
CEMS
as
an
option
to
periodic
stack
testing
and
parametric
monitoring
for
all
lean
burn
and
CI
engines
in
the
final
rule,
but
it
is
not
required.
Sources
that
are
using
CO
CEMS
to
comply
with
the
CO
percent
reduction
requirement
do
not
have
to
conduct
semiannual
performance
testing
for
CO.

V.
A.
2
Comment:
One
commenter
(
549)
suggested
that
EPA
eliminate
the
requirements
for
continuous
and
periodic
CO
monitoring
if
the
CO
concentration
in
the
exhaust
is
demonstrated
to
be
less
than
10
ppm
during
the
initial
performance
test.
Instrumental
CO
measurements
using
conventional
technology
can
be
made
reliably
at
CO
concentrations
as
low
as
approximately
10
ppm.
Below
this
concentration,
the
uncertainty
of
the
measured
value
becomes
large.
Carbon
monoxide
measurement
uncertainty
arises
from
the
cumulative
errors
associated
with
calibration
gas
accuracy,
analyzer
interferences
and
drift,
sampling
system
bias,
errors
associated
with
the
oxygen
measurement
needed
to
correct
CO
concentration
from
actual
to
15
percent
oxygen,
CO
concentration
gradients
in
the
stack,
and
other
factors.
The
concentration
of
CO
in
the
exhaust
from
some
engines
is
intrinsically
low
due
to
engine
design.
Future
improvements
in
engine
technology
are
likely
to
reduce
CO
emissions
even
further
than
present
engine
designs.
For
these
engines,
the
CO
concentration
after
an
oxidation
catalyst
system
achieving
93
percent
CO
reduction
could
be
below
10
ppm.
For
example,
record
#
29.41x
in
the
Emissions
Database
shows
as­
measured
CO
concentrations
of
126
ppm
in
the
engine
exhaust.
Ninety­
three
percent
reduction
from
this
level
corresponds
to
a
CO
concentration
of
approximately
9
ppm.
Compliance
data
in
this
situation
would
be
based
on
numbers
with
high
uncertainty,
thereby
increasing
the
risk
of
false
non­
compliance
to
unacceptable
levels.

Response:
The
EPA
agrees
with
the
commenter
that
some
but
not
all
of
the
cited
components
of
measurement
uncertainty
potentially
can
be
a
more
significant
percentage
at
lower
concentrations
of
CO
and
thereby
increase
the
risk
of
both
false
non­
compliance
and
false
compliance
decisions.
However,
EPA
does
not
agree
with
the
commenter
that
this
necessarily
requires
that
EPA
eliminate
measurement
requirements.
There
are
technology
and
techniques
capable
of
attaining
comparable
accuracy
and
precision.
The
EPA
has
recently
(
October
10,
2003)
proposed
changes
in
the
instrumental
methods
(
methods
10
and
3A
specifically
apply
here)
to
make
the
methods
performance
based
to
encourage
new
and
better
technology
use
and
to
calculate
an
upper
and
lower
uncertainty
range
reported
as
part
of
the
method.
The
EPA
has
also
proposed
multiple
point
sampling
to
account
for
potential
stack
stratification.
It
is
easier
to
get
a
bad
"
effect"
at
low
CO
concentrations,
oxygen
"
effects"
do
not
change.
63
V.
A.
3
Comment:
One
commenter
(
567)
was
confident
that
existing
monitoring
methodology
and
technology
should
be
able
to
accurately
measure
the
CO
emissions
after
control
technology
for
RICE.

Response:
The
EPA
has
noted
the
commenter's
input
and
agrees
with
the
commenter
that
there
is
technology
available
that
can
accurately
measure
10
ppm
or
less
CO.

V.
A.
4
Comment:
One
commenter
(
587)
observed
that
deficiencies
noted
in
the
proposed
rule
with
regard
to
the
test
methods
and
performance
protocols
render
CO
CEMS
infeasible
for
the
RICE
MACT.
While
CO
CEMS
have
been
demonstrated
on
some
facility
types,
their
application
to
RICE
is
very
limited.
Vendor
claims
for
CO
CEMS
and
CO
instrumental
analyzers,
unless
accompanied
by
emissions
test
data
obtained
under
known
and
controlled
conditions
applicable
to
the
subject
source
type,
should
not
be
considered
adequate
proof
of
availability
and
performance.
While
it
may
be
appropriate
for
EPA
to
solicit
comments
on
its
test
methods
and
technical
monitoring
requirements,
the
commenter
found
that
it
is
inappropriate
to
propose
requirements
for
measurement
systems
prior
to
resolving
the
current
deficiencies
with
the
EPA
protocols.

Response:
The
EPA
disagrees
with
the
commenter
that
the
application
of
CO
CEMS
must
be
considered
infeasible
for
all
RICE
unless
accompanied
by
emission
test
data
obtained
under
known
and
controlled
conditions
applicable
to
the
subject
source
category.
Since
EPA
has
previously
established
acceptable
CEMS
performance
specifications,
EPA
can
allow
the
RICE
source
owner
and
operator
the
optional
use
of
CO
CEMS
within
such
performance
standards
as
an
effective
parameter
monitor.
However,
as
discussed
above,
the
EPA
does
agree
that
it
should
not
require
the
installation
of
CEMS
at
all
affected
facilities.

V.
A.
5
Comment:
One
commenter
(
598)
stated
that
a
downstream
CEMS
should
be
sufficient,
but
only
when
a
NOx
CEMS
is
required.

Response:
The
EPA
is
no
longer
requiring
CO
CEMS,
therefore
the
comment
is
moot.

V.
A.
6
Comment:
One
commenter
(
605)
made
the
following
comment
regarding
CEMS:
Method
10
CO
CEMS
­
The
proposed
rule
calls
for
CEMS
to
measure
a
target
of
5
ppmv
CO
by
NDIR.
This
is
not
feasible
by
NDIR
but
is
achievable
by
Gas
Filter
Correlation
IR
(
GFC­
IR).
Source
testing
has
determined
a
de
minimus
level
of
2
ppm
with
this
technology.
Some
precautions
are
recommended:

(
1)
The
analyzer
must
be
calibrated
with
approximately
the
same
level
of
carbon
dioxide
(
CO
2)
as
the
source
because
there
is
some
overlap
of
CO
2
to
CO
(
percent
to
ppm).

(
2)
GFC­
IR
CEMS
accuracy
is
checked
by
reference
GFC­
IR.

(
3)
It
is
possible
to
confirm
accuracy
by
spiking
the
sample
up
to
the
regulatory
limit,
64
provided
that
the
spike
is
not
more
than
one­
tenth
of
the
sample.

Response:
The
EPA
has
noted
the
commenter's
concern
regarding
NDIR.
However,
Performance
Specifications
4,
4a,
and
4b
already
allow
any
type
of
CEMS
as
long
as
it
passes
the
performance
requirements.
Note
that
EPA
is
no
longer
requiring
lean
burn
and
CI
engines
5,000
HP
or
greater
to
demonstrate
compliance
with
the
CO
reduction
requirement
by
using
a
CO
CEMS.
The
use
of
CO
CEMS
to
demonstrate
compliance
is
optional
for
all
sources
meeting
the
CO
percent
reduction
requirement.

B.
Operating
Parameters
1.
All
Subcategories
V.
B.
1.1
Comment:
Four
commenters
(
542,
554,
556,
587)
felt
that
EPA
should
allow
owners
and
operators
the
opportunity
to
identify
more
appropriate
temperature
ranges
for
lean
burn
and
rich­
burn
engines
based
on
performance
testing,
control
device
design
specifications,
manufacturer
recommendations,
or
other
applicable
information.

Response:
The
EPA
feels
that
the
operating
parameters
it
has
specified
in
the
final
rule
for
sources
using
oxidation
catalyst
and
NSCR
are
appropriate
based
on
available
information
from
the
testing
at
CSU,
information
from
catalyst
vendors,
and
information
provided
in
comment
letters
submitted
to
the
docket.
Therefore,
EPA
has
not
made
the
change
recommended
by
the
commenters
in
the
final
rule.
Owners
and
operators
have
the
option
to
petition
the
Administrator
for
other
operating
parameters
following
the
procedures
in
section
63.8
of
the
General
Provisions
for
alternative
continuous
monitoring
system
(
CMS).
How
EPA
determined
the
temperature
ranges
for
lean
burn
engines
is
provided
in
response
to
comment
V.
B.
1.3.
Note
that
for
sources
that
are
meeting
the
emission
limitations
(
percent
reduction
requirement
or
formaldehyde
concentration
requirement)
of
the
final
rule
that
are
not
using
oxidation
catalyst
or
NSCR,
EPA
has
not
specified
which
operating
parameters
must
be
monitored
and
which
operating
limitations
these
sources
must
meet.
These
sources
must
petition
the
Administrator
for
approval
of
operating
limitations.

V.
B.
1.2
Comment:
Nine
commenters
(
542,
544,
552,
554,
556,
565,
578,
580,
587)
asserted
that
the
fuel
flow
and
HP
limits
should
be
removed.
Five
commenters
(
542,
554,
556,
580,
587)
recommended
that
EPA
specify
that
the
emission
standards
only
apply
within
a
60­
100
percent
load
range
and
performance
testing
should
be
conducted
within
that
load
range.
Commenter
544
suggested
revising
MACT
requirements
to
have
emission
limits
and
performance
testing
applicable
at
higher
load
conditions
instead
of
establishing
the
lowest
load
to
be
operated
in
the
future.
Another
commenter
(
578)
recommended
that
the
final
standards
only
apply
down
to
the
lowest
load
for
which
EPA
has
data
and
should
specify
that
the
performance
test
be
conducted
in
that
load
range.
Commenter
565
stated
that
should
EPA
pursue
minimum
load
testing
and
compliance
in
the
final
rule,
the
owner
and
operators
should
be
allowed
to
retest
the
unit
at
some
time
later
than
the
initial
performance
test
to
enlarge
the
operating
range.
The
lower
operating
load
and
fuel
range
should
then
be
based
on
the
lowest
load
that
has
demonstrated
65
compliance
irrespective
of
whether
the
demonstration
occurred
in
the
initial
or
later
performance
tests.

One
commenter
(
605)
stated
that
the
NESHAP
provide
two
options.
One
is
to
use
a
catalyst
and
the
other
is
to
limit
the
formaldehyde.
If
the
formaldehyde
limit
is
chosen,
however,
the
engine
must
maintain
an
operating
load
of
95
percent
or
more
of
the
load
established
in
the
initial
testing,
which
under
many
circumstances
is
impractical.
For
example,
this
option
cannot
be
chosen
for
the
commonly
used
variable­
load
application
engine.
For
variable
load
engines,
there
is
no
choice
but
to
use
a
catalyst.
The
commenter
stated
that
this
approach
limits
the
flexibility
in
controlling
these
engines.

Response:
In
the
proposed
rule,
EPA
required
sources
complying
with
the
alternative
formaldehyde
limit
to
maintain
an
operating
load
equal
to
or
greater
than
95
percent
of
the
operating
load
established
during
the
initial
performance
test
or
maintain
a
fuel
flow
rate
equal
to
or
greater
than
95
percent
of
the
fuel
flow
rate
established
during
the
initial
performance
test.
These
sources
were
also
required
to
comply
with
any
additional
operating
limitations
approved
by
the
Administrator.
Based
on
information
received
during
the
public
comment
period,
EPA
has
reached
the
conclusion
that
maintaining
the
load
or
fuel
flow
rate
within
95
percent
of
that
established
during
the
initial
performance
test
may
be
impractical
for
many
applications,
especially
those
in
load
following
applications.
Therefore,
the
EPA
has
not
included
the
requirement
to
maintain
load
or
fuel
flow
rate
in
the
final
rule.

In
the
final
rule,
EPA
has
allowed
sources
the
flexibility
of
complying
with
either
emission
limitation
requirement,
regardless
of
which
control
technique
is
applied.
Sources
that
use
an
oxidation
catalyst
are
not
limited
to
meeting
the
requirement
to
reduce
CO
emissions,
but
have
the
option
to
comply
with
either
the
CO
percent
reduction
requirement
or
with
the
formaldehyde
concentration
limit.
Similarly,
4SRB
engines
that
use
NSCR
are
not
limited
to
meeting
the
requirement
to
reduce
formaldehyde
emissions,
but
have
the
option
to
comply
with
either
the
formaldehyde
percent
reduction
requirement
or
with
the
formaldehyde
concentration
limit.
Sources
that
do
not
use
oxidation
catalyst
or
NSCR,
may
choose
to
comply
with
either
the
percent
reduction
requirement
or
with
the
formaldehyde
concentration
limit.

Sources
complying
with
the
alternative
formaldehyde
limit
that
use
oxidation
catalysts
or
NSCR
must
continuously
monitor
and
maintain
the
catalyst
inlet
temperature
and
measure
the
catalyst
pressure
drop
monthly.
Sources
complying
with
the
alternative
formaldehyde
limit
that
do
not
use
oxidation
catalysts
or
NSCR
must
petition
the
Administrator
for
operating
limitations
to
be
continuously
monitored.
In
the
petition
for
approval
of
operating
limitations,
EPA
recommends
that
sources
consider
establishing
load
or
fuel
flow
rate
as
possible
operating
parameters
to
continuously
monitor.
Finally,
the
EPA
has
based
the
emission
standard
on
test
results
from
high
load
tests
only.
Typically,
as
load
decreases,
the
concentration
of
HAP
increases.
Comments
received
support
this
trend.
Therefore,
EPA
has
specified
in
the
final
rule
that
performance
tests
must
be
conducted
at
high
load
conditions,
defined
as
100
percent
±
10
percent.

V.
B.
1.3
Comment:
Seven
commenters
(
542,
549,
554,
556,
571,
580,
587)
contended
66
that
the
temperature
ranges
at
the
catalyst
inlet
should
be
revised.
Six
commenters
(
542,
549,
554,
556,
580,
587)
supported
an
operating
range
of
450

F­
1,350

F
for
lean
burn
engines
and
the
ability
to
develop
customized
catalyst
inlet
temperature
ranges
based
on
specific
engine
operating
parameters.
Commenter
571
recommended
using
450

F
minimum
catalyst
inlet
temperature
for
2SLB.
Commenter
587
also
said
that
owners
and
operators
should
be
allowed
to
identify
more
appropriate
temperature
ranges
based
on
performance
testing,
control
device
design
specifications,
manufacturer
recommendations,
or
other
applicable
information
(
such
as
a
performance
test
on
a
similar
unit).

Response:
The
EPA
proposed
that
lean
burn
and
CI
engines
complying
with
the
requirement
to
reduce
CO
emissions
maintain
the
temperature
of
the
stationary
RICE
exhaust
so
that
the
catalyst
inlet
temperature
is
greater
than
or
equal
to
500

F
and
less
than
or
equal
to
1,250

F.
The
EPA
required
the
catalyst
inlet
temperature
to
be
maintained
to
ensure
proper
operation
of
the
oxidation
catalyst.
The
EPA
stated
in
the
preamble
to
the
proposed
rule
that,
in
general,
the
oxidation
catalyst
performance
will
decrease
as
the
catalyst
inlet
temperature
decreases.
Also,
if
the
catalyst
inlet
temperature
is
too
high,
oxidation
catalyst
performance
could
be
affected.
Finally,
the
oxidation
catalyst
inlet
temperature
cannot
be
too
low,
or
the
reduction
of
HAP
emissions
may
be
compromised.
For
these
reasons,
EPA
proposed
that
sources
complying
with
the
CO
reduction
requirement
using
an
oxidation
catalyst
maintain
the
catalyst
inlet
temperature
within
500­
1,250

F.
Several
comments
received
during
the
public
comment
period
indicated
that
the
temperature
range
EPA
proposed
for
catalyst
inlet
temperature
should
be
expanded.
Commenters
suggested
that
the
lower
end
of
the
temperature
range
should
start
at
450

F.
The
level
of
the
standard
for
2SLB
engines
is
58
percent
CO
reduction.
Similar
CO
reduction
was
seen
at
CSU
for
2SLB
engines
where
the
exhaust
temperature
was
450

F.
For
this
reason,
the
EPA
agrees
with
the
commenters
that
the
catalyst
inlet
lower
temperature
should
be
set
at
450

F.
Furthermore,
EPA
feels
that
the
oxidation
catalyst
will
perform
adequately
at
a
temperature
of
1,350

F.
This
was
discussed
in
the
memorandum
entitled
"
Control
Technologies
for
Stationary
RICE"
available
from
EPA's
Edocket
Website
as
Document
ID
Number
OAR­
2002­
0059­
0064
(
A­
95­
35
II­
B­
41).
Commenters
also
stated
that
Waukesha
Pearce
Industries,
Inc.
includes
1,350

F
in
their
limited
warranty
statements
for
oxidation
catalysts.
Therefore,
the
EPA
has
written
the
temperature
range
requirement
for
catalyst
inlet
temperatures
to
be
between
450

F
and
1,350

F
in
the
final
rule.
Regarding
the
comment
that
owners
and
operators
should
be
allowed
to
identify
more
appropriate
temperature
ranges,
EPA
feels
that
requiring
a
catalyst
inlet
temperature
range
of
450

F
to
1,350

F
is
appropriate.
Based
on
information
from
the
testing
at
CSU,
information
from
catalyst
vendors,
and
information
provided
in
comment
letters
submitted
to
the
docket,
EPA
feels
it
has
adequate
information
that
supports
requiring
a
catalyst
inlet
temperature
range
of
450

F
to
1,350

F,
and
EPA
does
not
feel
it
is
necessary
to
allow
owners
and
operators
the
ability
to
identify
and
define
other
temperature
ranges.
Owners
and
operators
have
the
option
to
petition
the
Administrator
for
other
operating
parameters
following
the
procedures
in
section
63.8
for
alternative
monitoring
procedures.

V.
B.
1.4
Comment:
One
commenter
(
598)
expressed
that
the
temperature
measurements
for
lean
burn
engines
are
acceptable
as
proposed
because
they
do
not
specify
a
minimum
temperature
increase,
only
that
the
catalyst
is
hot
enough
to
work,
and
not
too
hot
to
be
67
damaged.

Response:
The
EPA
has
noted
the
commenter's
statement
regarding
temperature
measurements
for
lean
burn
engines
and
agrees
generally
with
the
comments
though
EPA
is
revising
them
to
some
extent
to
respond
to
the
comments
discussed
in
V.
B.
1.3.

V.
B.
1.5
Comment:
Eight
commenters
(
542,
549,
552,
556,
571,
580,
587,
598)
stated
that
the
requirement
to
measure
pressure
drop
should
be
removed.
Commenter
571
indicated
that
the
operating
limitation
not
to
exceed
a
pressure
change
of
2
inches
of
water
column
from
the
initial
performance
test
has
the
potential
to
be
problematic
in
practice.
The
intent
by
EPA
is
to
designate
an
easily
measurable
quantity
that
will
give
an
indication
"...
if
the
oxidation
catalyst
is
damaged
or
fouled..."
Engines
in
the
field
vary
speed
and
load
as
needed
to
perform
their
tasks.
These
variations
will
change
the
flow
through
the
catalyst
and,
accordingly,
change
the
pressure
drop.
Unless
the
initial
performance
test
maps
the
entire
speed
and
load
operating
envelope
and
a
suitable
sophisticated
controller
is
used
to
monitor
the
pressure
drop
and
compare
it
to
the
map
at
every
operating
point
false
deviations
are
likely.
Another
commenter
(
598)
stated
that
there
is
no
need
for
continuous
pressure
drop
measurements
on
engines
running
exclusively
on
natural
gas
and
at
high
loads.
The
commenter
has
seen
very
little
problems
with
catalyst
fouling
on
their
lean
burn
RICE
equipped
with
oxidation
catalysts.
The
commenter
understood
that
it
is
an
issue
in
some
installations,
but
concludes
that
they
would
be
applications
either
running
on
other
fuels
or
where
engines
are
run
at
idle
or
very
low
load
for
long
periods
of
time.
Commenter
587
stated
that
the
proposed
requirements
to
continuously
monitor
and
maintain
a
prescribed
pressure
differential
across
the
catalyst
should
be
removed
from
the
final
rule
for
the
following
reasons:

(
1)
Although
significant
change
in
differential
pressure
across
the
catalyst
may
provide
an
indication
that
the
catalyst
has
become
fouled,
EPA
has
presented
no
evidence
to
suggest
that
an
increase
in
2
inches
of
water
column
means
that
catalyst
performance
is
impacted;

(
2)
Industry
data
demonstrates
that
the
pressure
drop
can
increase
more
than
2
inches
of
water
column
without
impacting
catalyst
performance.
Such
increases
may
even
occur
because
of
engine
operating
conditions.
For
that
reason,
EPA's
proposed
2
inches
of
water
column
condition
might
forbid
engines
to
operate
within
part
of
their
normal
operating
range;
and
(
3)
Vendors
do
not
treat
pressure
differential
as
a
continuous
operating
parameter
requirement.
Rather
it
is
presented
as
a
maintenance
requirement
for
catalysts
on
some
engines.
The
general
duty
clause
of
§
63.6(
e)(
1)(
i)
is
sufficient
to
address
pressure
drop
issues.

Finally,
commenter
598
stated
that
the
uniqueness
of
the
installation
should
be
given
consideration
in
whether
or
not
pressure
drop
is
required
to
be
monitored.

Response:
The
EPA
proposed
a
requirement
for
4SRB
engines
complying
with
the
68
requirement
to
reduce
formaldehyde
emissions
using
NSCR
and
2SLB,
4SLB,
and
CI
engines
less
than
5,000
HP
complying
with
the
requirement
to
reduce
CO
emissions
using
an
oxidation
catalyst
to
maintain
the
catalyst
so
that
the
pressure
drop
across
the
catalyst
does
not
change
by
more
than
2
inches
of
water
from
the
pressure
drop
across
the
catalyst
measured
during
the
initial
performance
test.
Catalyst
vendors
have
indicated
to
EPA
that
the
pressure
drop
across
the
catalyst
may
be
a
good
parameter
to
indicate
catalyst
performance
and
that
an
increase
in
pressure
drop
is
an
indication
of
poor
catalyst
performance.
The
pressure
drop
across
the
catalyst
can
indicate
if
the
catalyst
is
damaged
or
fouled.
If
the
catalyst
is
damaged
or
becomes
fouled,
the
catalyst
performance
would
decrease.
For
the
reasons
provided,
EPA
feels
it
is
appropriate
to
use
the
pressure
drop
as
it
serves
as
a
surrogate
of
the
catalyst
performance.

The
EPA
determined
at
proposal
that
if
the
pressure
drop
across
the
catalyst
deviates
by
more
than
2
inches
of
water
from
the
pressure
drop
across
the
catalyst
measured
during
the
initial
performance
test,
the
catalyst
might
be
damaged
or
fouled.
This
was
based
on
information
received
from
catalyst
vendors
which
indicated
that
if
the
pressure
drop
changes
by
more
than
2
inches
of
water
column,
the
catalyst
should
be
inspected
for
damage
or
fouling.
For
this
reason,
EPA
feels
it
was
appropriate
to
specify
that
the
pressure
drop
across
the
catalyst
should
not
change
by
more
than
2
inches
from
the
pressure
drop
measured
during
the
initial
performance
test.
Anything
higher
than
2
inches
might
indicate
damage
or
fouling
of
the
catalyst.
The
EPA
feels
it
is
appropriate
to
maintain
the
pressure
drop
requirement
as
proposed.
However,
the
EPA
has
reevaluated
its
position
regarding
requiring
sources
to
monitor
the
pressure
drop
across
the
catalyst
on
a
continuous
basis
and
is
no
longer
requiring
sources
to
install
a
CPMS
to
monitor
this
parameter
continuously.
The
pressure
drop
across
the
catalyst
is
not
likely
to
change
within
short
periods
of
time,
but
is
a
parameter
the
owner
or
operator
might
see
changing
over
a
longer
period
of
time,
not
within
hours
or
days.
This
is
consistent
with
comments
that
stated
that
vendors
do
not
treat
pressure
differential
as
a
continuous
operating
parameter
requirement.
Rather
it
is
presented
as
a
maintenance
requirement
for
catalysts
on
some
engines.
For
this
reason,
EPA
feels
it
is
appropriate
to
require
sources
that
must
comply
with
the
pressure
drop
requirement
to
measure
this
parameter
monthly,
as
EPA
does
not
expect
the
pressure
drop
across
the
catalyst
to
change
significantly
more
frequently
than
monthly.
Regarding
the
comment
that
the
uniqueness
of
the
installation
should
be
given
consideration
in
whether
or
not
pressure
drop
is
required
to
be
monitored,
EPA
feels
that
it
has
gathered
sufficient
information
from
catalyst
vendors
that
supports
requiring
the
pressure
drop
to
be
monitored
and
maintained
monthly.
In
addition,
the
commenter
did
not
describe
or
provide
information
regarding
how
the
uniqueness
of
the
installation
would
affect
whether
or
not
monitoring
and
maintaining
the
pressure
drop
should
be
required.

Regarding
the
comment
that
engines
in
the
field
vary
speed
and
load
as
needed
to
perform
their
tasks
and
that
these
variations
will
change
the
flow
through
the
catalyst
and,
accordingly,
change
the
pressure
drop,
EPA
contacted
the
commenter
for
clarification
as
EPA
was
unclear
on
how
speed
could
affect
the
pressure
drop.
The
commenter
stated
that
by
changing
the
speed
and
load
of
an
engine,
both
the
exhaust
flow
rate
and
the
temperature
of
that
flow
will
change.
The
commenter
provided
the
example
that
if
you
reduce
load
on
the
engine
by
25
percent
it
will
need
25
percent
less
air
and
25
percent
less
fuel
(
all
other
things
being
constant).
The
commenter
said
69
that
this
means
that
the
exhaust
flow
is
also
reduced
by
25
percent.
The
commenter
continued
by
stating
that
lower
loading
means
that
the
engine
is
working
less
hard
so
the
temperature
of
this
reduced
gas
stream
is
also
somewhat
lower.
With
this
lower
flow
and
reduced
temperature
through
the
fixed
restriction
it
means
that
the
pressure
drop
across
the
catalyst
will
change
(
drop
in
this
case).
The
reverse
happens
for
a
load
increase.
The
commenter
noted
that
a
speed
change
has
similar
results.
The
commenter
added
that
any
time
the
engine
changes
its
operating
speed
or
load,
the
pressure
drop
across
the
catalyst
will
change
totally
independent
of
aging
or
dirt.
The
commenter's
main
concern
when
submitting
the
comments
summarized
here
was
that
setting
a
certain
pressure
drop
as
an
indication
of
the
need
to
replace
or
service
a
catalyst
will
not
work
as
a
load
or
speed
increase
can
also
result
in
the
same
2
inches
of
water
increase
in
pressure
drop
across
the
catalyst.
The
commenter
stated
that
catalysts
vary
considerably
in
size
and
shape
so
there
may
not
be
a
"
one
size
fits
all"
pressure
increase
that
works
well
everywhere.

The
EPA
understands
the
commenter's
concern.
However,
EPA
investigated
several
parameters
and
looked
at
different
options
for
establishing
operating
limitations
and
determined
based
on
information
received
from
catalyst
vendors
that
the
pressure
drop
across
the
catalyst
is
an
appropriate
indicator
of
catalyst
performance.
The
EPA
acknowledges
that
the
pressure
drop
across
the
catalyst
might
be
greater
than
2
inches
compared
to
the
pressure
drop
across
the
catalyst
measured
during
the
initial
performance
test,
yet
the
catalyst
may
still
be
functioning
properly.
However,
EPA's
intent
is
for
sources
to
measure
the
pressure
drop
monthly
and
if
the
pressure
drop
is
beyond
the
established
ranges,
sources
should
inspect
the
catalyst
to
ensure
that
it
is
operating
properly.
Sources
must
report
periods
when
operating
parameters
stray
from
the
established
operating
limitations
or
ranges
in
their
semiannual
compliance
report.
However,
straying
from
established
operating
limitations
is
not
considered
a
violation,
in
and
of
itself.
Measuring
the
catalyst
pressure
drop
monthly
is
intended
to
provide
sources
with
a
chance
to
determine
if
the
catalyst
is
still
operating
properly
and
an
opportunity
to
replace,
service,
wash,
etc.
the
catalyst,
if
necessary.
If
the
pressure
drop
exceeds
2
inches
of
water
compared
to
the
pressure
drop
established
during
the
initial
performance
test,
sources
would
not
necessarily
be
required
to
replace
or
service
the
catalyst,
but
would
be
required
to
determine
the
causes
for
the
increased
pressure
drop
and
demonstrate
that
the
change
in
pressure
drop
beyond
the
acceptable
ranges
is
due
to
other
factors
and
not
due
to
clogging
or
damage
of
the
catalyst
and
that
the
catalyst
is
still
operating
properly.
The
causes
for
the
increase
in
pressure
drop
beyond
the
range
established
during
the
initial
performance
test
should
be
documented
in
the
semiannual
compliance
report.
Parameter
deviations
are
not
direct
violations.
However,
if
for
example
the
operating
limitation
was
exceeded
2
or
more
months
in
a
row,
the
enforcing
agency
could
request
that
a
performance
test
be
conducted.

V.
B.
1.6
Comment:
One
commenter
(
570)
stated
that
Table
2b
of
the
proposed
rule
contains
requirements
for
2SLB,
4SLB,
and
CI
engines
under
5,000
HP
that
employ
an
oxidation
catalyst
to
reduce
CO.
The
table
does
not
specify
a
minimum
temperature
rise
across
the
catalyst.
The
commenter
said
that
it
should
specify
such
a
minimum
temperature
rise.

Response:
The
EPA
is
not
requiring
2SLB,
4SLB,
and
CI
engines
under
5,000
HP
that
use
an
oxidation
catalyst
to
maintain
the
temperature
rise
across
the
catalyst
within
a
certain
70
range.
In
the
proposal,
4SRB
engines
using
NSCR
were
required
to
maintain
the
temperature
rise
across
the
catalyst
within
5
percent
of
the
temperature
rise
across
the
catalyst
measured
during
the
initial
performance
test
(
in
addition
to
maintaining
the
catalyst
pressure
drop
and
catalyst
inlet
temperature).
The
requirement
to
maintain
the
temperature
rise
across
the
catalyst
for
4SRB
engines
has
been
removed
from
the
final
rule.
New
2SLB,
4SLB,
and
CI
engines
that
use
an
oxidation
catalyst
must:
(
1)
maintain
the
catalyst
so
that
the
pressure
drop
across
the
catalyst
does
not
change
by
more
than
2
inches
of
water
from
the
pressure
drop
across
the
catalyst
that
was
measured
during
the
initial
performance
test;
and
(
2)
maintain
the
temperature
of
the
stationary
RICE
exhaust
so
that
the
catalyst
inlet
temperature
is
greater
than
or
equal
to
450

F
and
less
than
or
equal
to
1,350

F.

V.
B.
1.7
Comment:
Two
commenters
(
526,
579)
requested
that
standards
should
allow
sources
to
establish
all
appropriate
operating
limits
based
on
operating
data
collected
through
the
initial
catalyst
life
period,
up
to
5
years,
so
that
truly
representative
engine­
in­
the­
field
data
can
be
obtained.

Response:
The
EPA
disagrees
with
the
commenters.
Owners
and
operators
must
use
the
operating
parameters
that
have
been
specified
in
the
rule.
However,
EPA
has
provided
some
relief
in
that
the
requirement
to
monitor
and
maintain
the
operating
load
or
fuel
flow
rate
for
sources
complying
with
the
formaldehyde
concentration
emission
limitation
has
not
been
included
in
the
final
rule.
The
rationale
for
this
was
provided
in
response
to
comment
V.
B.
1.2.
In
addition,
the
requirement
for
4SRB
engines
complying
with
the
requirement
to
reduce
formaldehyde
emissions
to
maintain
the
temperature
rise
across
the
catalyst
has
not
been
included
in
the
final
rule.
The
reasons
for
excluding
this
requirement
from
the
final
rule
was
discussed
in
response
to
comment
V.
B.
2.1.
As
far
as
the
remaining
operating
limits
EPA
has
included
in
the
final
rule,
EPA
feels
that
the
information
in
the
docket
supports
the
catalyst
inlet
temperature
ranges
and
pressure
drop
that
EPA
has
specified.
This
is
discussed
further
in
response
to
comments
V.
B.
1.3
and
V.
B.
1.5.
For
these
reasons,
EPA
feels
the
operating
limits
of
the
final
rule
are
appropriate
and
does
not
feel
it
is
necessary
to
allow
sources
to
establish
their
own
operating
limits.
Note
that
in
the
final
rule,
EPA
is
only
specifying
operating
limitations
for
sources
that
use
oxidation
catalysts
or
NSCR.
Sources
that
do
not
use
oxidation
catalysts
or
NSCR
must
petition
the
Administrator
for
approval
of
operating
limitations
or
approval
or
no
operating
limitations.

V.
B.
1.8
Comment:
Five
commenters
(
549,
551,
565,
569,
571)
expressed
that
it
is
unnecessary
to
reestablish
operating
limits
when
catalysts
are
replaced.
One
commenter
(
551)
stated
that
the
requirement
in
§
63.6640(
b)
to
retest,
report
and
reestablish
operating
limitations
is
not
necessary
unless
there
has
been
a
system
redesign
or
change
in
catalyst
type.
Another
commenter
(
571)
indicated
that
reestablishing
limits
is
not
necessary
when
the
catalyst
is
changed
for
an
identical
replacement
and
this
requirement
should
be
deleted.
Two
other
commenters
(
565,
569)
stated
that
the
same
design
catalyst
obtained
from
the
same
supplier
will
provide
the
same
performance
under
the
same
operating
conditions.

Response:
The
EPA
disagrees
with
the
commenters
and
feels
it
is
important
to
reestablish
operating
limitations.
The
enforcement
agency
needs
confirmation
of
the
performance
of
the
71
replacement
catalyst.
The
EPA
does
not
think
that
retesting
and
reestablishing
operating
limitations
will
occur
frequently,
and
therefore
does
not
think
this
will
be
a
significant
burden
for
facilities.
This
is
also
typically
required
by
the
States.

V.
B.
1.9
Comment:
One
commenter
(
569)
stated
that
operating
parameters
established
during
initial
performance
tests
should
not
become
permit
limits.

One
commenter
(
585)
was
opposed
to
EPA's
use
of
performance
tests
to
define
operating
limits
for
each
RICE
unit.
The
performance
test
should
only
be
used
to
determine
if
a
unit
can
meet
the
standard.

One
commenter
(
569)
remarked
that
periods
when
indicator
parameters
stray
from
the
established
ranges
are
not
violations.

One
commenter
(
569)
said
that
all
references
to
a
continuous
parameter
monitoring
system
(
CPMS)
in
Table
5
of
the
proposed
rule
and
all
references
to
data
obtained
from
CPMS
in
Table
6
of
the
proposed
rule
are
unnecessary
and
should
be
deleted.

One
commenter
(
569)
stated
that
all
references
to
"
Operating
Limitations"
should
be
replaced
with
"
Operator
Indicator
Parameters"
in
Tables
1B
and
2B
of
the
proposed
rule.

One
commenter
(
570)
noted
that
the
rule
requires
owners
and
operators
who
comply
with
the
formaldehyde
limit
to
petition
the
Administrator
prior
to
conducting
the
initial
performance
test
(
which
must
be
conducted
within
180
days).
The
commenter
objects
to
the
rule
making
a
facility
reliant
on
the
actions
of
the
Administrator
in
such
a
short
time
window.
The
commenter
suggested
that
the
rule
specify
what
additional
parameters
are
to
be
limited
or
monitored
within
the
rule
or
allow
owners
and
operators
the
flexibility
to
establish
them
on
their
own
accord.

One
commenter
(
565)
contended
that
due
to
periodic
performance
testing,
on­
going
operating
parameter
limits
are
not
necessary
and
should
be
deleted.

One
commenter
(
564)
expressed
that
operating
limitations
that
relate
to
catalyst
pressure
drop
and
temperature
differential
have
the
practical
effect
of
beyond­
the­
floor
requirements
that
are
not
justified.
In
addition,
the
Agency
presents
no
information
to
suggest
that
the
pressure
drop
prescribed
indicates
that
the
catalyst
has
become
fouled
to
the
point
of
creating
unacceptable
catalyst
performance.
Similarly,
the
Agency
presents
no
information
to
suggest
that
the
allowable
temperature
differential
prescribed
is
a
good
indicator
of
catalyst
performance
and
that
temperature
differentials
outside
of
the
range
result
in
unacceptable
catalyst
performance.

Response:
The
EPA
disagrees
with
these
comments.
The
EPA's
intent
is
to
ensure
that
emission
limitations
are
continuously
being
met
and
has
included
requirements
to
monitor
and
maintain
certain
operating
parameters
between
periodic
stack
tests.
Therefore,
periods
when
operating
parameters
stray
from
the
established
ranges
are
considered
deviations
that
need
to
be
reported
to
EPA
semiannually.
As
discussed
in
response
to
comment
V.
B.
1.2,
EPA
has
removed
72
the
load
or
fuel
flow
rate
requirement
for
sources
complying
with
the
alternative
formaldehyde
concentration
emission
limitation.
The
EPA
has
also
removed
the
requirement
to
continuously
monitor
and
maintain
the
pressure
drop
across
the
catalyst.
The
pressure
drop
across
the
catalyst
must
be
measured
monthly.
This
was
discussed
in
response
to
comment
V.
B.
1.5.
The
EPA
has
also
removed
the
requirement
to
continuously
monitor
the
temperature
rise
across
the
catalyst
for
4SRB
engines
complying
with
the
requirement
to
reduce
formaldehyde.
Regarding
the
comment
that
operating
parameters
established
during
the
initial
performance
test
should
not
become
permit
limits,
this
decision
is
up
to
the
States
and
is
outside
the
scope
of
this
rule.
Owners
and
operators
that
stray
from
the
operating
limitations
of
the
final
rule,
must
report
these
as
deviations.
However,
a
deviation
is
not
a
violation,
but
might
result
in
further
examination
by
the
enforcing
agency.
The
EPA
feels
that
it
has
provided
sufficient
time
for
sources
complying
with
the
formaldehyde
concentration
emission
limitation
that
do
not
use
oxidation
catalysts
or
NSCR
to
petition
the
Administrator.
The
EPA
is
unable
to
establish
operating
limitations
for
sources
complying
with
the
formaldehyde
concentration
limit
that
do
not
use
oxidation
catalysts
or
NSCR
because
it
does
not
know
which
control
technique
(
if
any)
sources
will
apply
to
meet
the
formaldehyde
limit
and
therefore
does
not
know
which
parameters
relate
to
and
can
be
used
as
surrogates
for
the
control
device's
performance.

2.
4SRB
V.
B.
2.1
Comment:
Eight
commenters
(
542,
549,
552,
556,
571,
580,
587,
598)
stated
that
the
requirement
to
measure
the
temperature
rise
for
rich
burn
RICE
should
be
removed.
Commenter
571
had
the
opinion
that
5
percent
difference
in
temperature
is
not
feasible
or
workable
in
practice.
While
a
NSCR
catalyst
is
more
likely
to
show
a
positive
temperature
change
across
the
catalyst,
very
low,
or
even
negative,
temperature
changes
are
possible
while
the
catalyst
is
functioning
normally.
Commenter
598
did
not
think
it
is
appropriate
to
specify
that
the
temperature
rise
across
a
NSCR
catalyst
has
to
stay
within
5
percent
of
the
temperature
rise
(
or
any
other
specific
value)
measured
at
the
initial
source
test.
The
commenter
stated
that
this
seems
arbitrary.
At
one
facility,
the
commenter
has
seen
zero
temperature
change
across
the
catalyst.
Yet,
NOx,
CO
and
volatile
organic
compounds
(
VOC)
reductions
were
all
occurring
at
high
efficiency
and
in
full
compliance
with
requirements.
It
would
be
more
appropriate
to
simply
require
that
NSCR
be
operated
in
conjunction
with
an
air/
fuel
ratio
controller
and
that
the
catalyst
inlet
temperature
simply
be
hot
enough
to
ensure
it
is
working,
but
not
too
hot
to
damage
the
catalyst.

One
commenter
(
570)
said
that
Table
1b
of
the
proposed
rule
stipulates
that
4SRB
RICE
must
ensure
that
the
temperature
rise
across
the
catalyst
is
no
more
than
5
percent
different.
The
commenter
asked
what
if
the
temperature
is
10
percent
different
and
would
this
not
represent
a
higher
degree
of
oxidation.
The
commenter
questioned
why
this
should
not
be
allowed.

Response:
As
summarized
above,
EPA
received
several
comments
regarding
the
requirement
in
the
proposed
rule
that
4SRB
engines
monitor
and
maintain
the
temperature
rise
across
the
NSCR.
Based
on
the
information
received,
EPA
agrees
with
the
commenters
that
such
a
requirement
would
be
inappropriate
and
most
likely
would
not
provide
an
accurate
73
representation
of
how
the
catalyst
is
performing.
The
EPA
is
including
the
requirement
to
measure
the
catalyst
pressure
drop
monthly
and
to
maintain
and
continuously
monitor
the
catalyst
inlet
temperature
to
ensure
that
it
remains
between
750

F
and
1,250

F.
It
is
EPA's
opinion
that
monitoring
and
maintaining
these
two
parameters
is
sufficient
to
ensure
proper
catalyst
operation.
Therefore,
EPA
has
not
included
the
requirement
to
maintain
the
catalyst
such
that
the
temperature
rise
across
the
catalyst
stays
within
5
percent
of
the
temperature
rise
measured
during
the
initial
performance
test
in
the
final
rule.

V.
B.
2.2
Comment:
Six
commenters
(
526,
542,
552,
556,
580,
587)
asked
that
EPA
allow
the
use
of
NOx
monitoring
as
a
surrogate
for
NSCR
function.
Commenter
552
stated
that
CO
monitoring
should
be
allowed.
Commenter
552
stated
that
the
measurement
of
formaldehyde
is
a
very
expensive
and
potentially
unreliable
undertaking
that
is
difficult
to
perform
under
field
conditions.
The
available
monitoring
methods,
such
as
Method
320
using
FTIR,
are
intended
for
use
by
highly
qualified
individuals
under
controlled
laboratory
conditions
and
may
not
be
suitable
to
all
field
conditions
where
RICE
operate.
Consequently,
to
adequately
address
these
technical
difficulties
and
the
high
costs
of
formaldehyde
testing,
EPA
needs
to
provide
a
compliance
option
or
alternative
for
rich
burn
engines.
Non­
selective
catalytic
reduction
has
a
long
history
of
successful
operation
to
reduce
NOx
and
CO
emissions,
and
monitoring
of
those
parameters
is
currently
used
to
ensure
that
the
catalyst
is
working
properly.
Since
EPA
has
determined
that
NSCR
is
the
best
available
technology
to
reduce
formaldehyde
and
other
HAP
from
rich
burn
engines,
it
follows
that
any
monitoring
technique
that
can
demonstrate
the
proper
operation
of
the
catalyst
should
be
sufficient
to
assure
that
corresponding
formaldehyde
reductions
also
are
occurring.
Although
there
currently
may
be
insufficient
data
to
adequately
determine
a
conclusive
and
linear
relationship
between
CO
and
NOx
reduction
with
formaldehyde
reductions,
commenter
552
knows
of
no
reason
why
CO
or
NOx
reduction
should
not
be
an
acceptable
surrogate
for
rich
burn
engines.

Response:
The
EPA
agrees
that
emission
testing
to
measure
HAP
emitted
from
engines
is
expensive.
As
a
result
EPA
investigated
using
CO
as
a
surrogate
for
HAP
emissions
as
CO
is
easy
to
measure.
In
addition,
equipment
to
measure
CO
emissions
is
readily
available.
However,
EPA
determined
that
there
is
not
a
good
relationship
between
CO
emissions
reductions
and
HAP
emissions
reductions
for
4SRB
engines
using
NSCR
systems.
For
example,
in
certain
cases,
the
CO
percent
reduction
efficiency
increased
while
the
formaldehyde
reduction
efficiency
decreased.
Please
refer
to
the
memorandum
entitled
"
CO
Removal
Efficiency
as
a
Surrogate
for
HAP
Removal
Efficiency
­
Spark
Ignition
Lean
Burn
Clean
Gaseous
Fueled
Engines
and
Compression
Ignition
Engines
Using
an
Oxidation
Catalyst"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35))
for
this
analysis.
Therefore,
EPA
found
it
inappropriate
to
use
CO
as
a
surrogate
for
HAP
emissions
for
4SRB
engines.
Emissions
data
available
to
EPA
does
not
support
allowing
NOx
monitoring
as
a
surrogate
for
NSCR
function
as
the
necessary
relationship
required
to
support
this
allowance
was
not
identified.
In
other
words,
EPA
could
not
establish
that
there
is
a
good
relationship
between
NOx
emissions
reductions
and
HAP
emissions
reductions.
One
commenter
even
stated
that
there
currently
may
be
insufficient
data
to
adequately
determine
a
conclusive
and
linear
relationship
between
CO
and
NOx
reduction
with
formaldehyde
reductions.
The
EPA
does
not
feel
it
is
appropriate
to
allow
NOx
monitoring
as
a
74
surrogate
for
NSCR
function.
Emissions
data
in
the
docket
show
that
even
though
NOx
decreases,
formaldehyde
and
HAP
emissions
may
increase,
which
could
lead
to
false
compliance
if
EPA
were
to
use
NOx
monitoring
as
a
surrogate
for
NSCR
function.

V.
B.
2.3
Comment:
One
commenter
(
526)
indicated
that
many
existing
4SRB
units
may
not
be
able
to
meet
the
traverse
requirements
in
Table
4
of
the
proposed
rule
if
there
is
not
sufficient
length
of
exhaust
piping
before
the
NSCR
catalyst.

Response:
These
sources
will
need
to
apply
for
an
alternative
to
the
method
requirement.
The
EPA
feels
the
traverse
requirements
are
appropriate.
The
traverse
requirements
are
needed
to
assure
accurate
measurements
taking
into
account
the
potential
for
stratification.

V.
B.
2.4
Comment:
Two
commenters
(
554,
578)
stated
that
EPA
should
allow
owners
and
operators
to
identify
more
appropriate
inlet
temperature
ranges
for
NSCR
catalysts
based
on
performance
testing,
control
device
design
specifications,
manufacturer
recommendations,
etc.
Commenter
578
also
stated
that
the
assumption
that
fixed
temperature/
pressure
drop
operating
ranges
can
be
applied
to
all
4SRB
engines
requires
reexamination.

Response:
The
EPA
has
specified
inlet
temperature
ranges
for
NSCR
catalysts
based
on
information
received
from
catalyst
vendors
and
feels
these
are
appropriate
as
proposed,
i.
e.,
750

F­
1,250

F.
Owners
and
operators
of
stationary
RICE
equipped
with
NSCR
have
the
option
to
petition
the
Administrator
for
alternative
inlet
temperature
ranges
provided
that
these
ranges
will
indicate
proper
catalyst
performance.
The
EPA
has
removed
the
requirement
to
maintain
and
continuously
monitor
the
temperature
rise
across
the
catalyst
from
the
final
rule.
This
was
discussed
in
response
to
comment
V.
B.
2.1.
The
EPA
has
kept
the
requirement
to
measure
the
pressure
drop
across
the
catalyst,
but
has
made
this
a
monthly
requirement
and
not
a
continuous
measurement.
This
was
discussed
in
response
to
comment
V.
B.
1.5.

C.
Initial
Notification
V.
C.
1
Comment:
Four
commenters
(
540,
551,
569,
585)
recommended
that
EPA
eliminate
the
initial
notification
requirement
for
emergency/
limited
use
engines
as
it
seems
unnecessary.
Commenter
551
expressed
that
this
is
duplicate
of
title
V
requirements.
Another
commenter
(
569)
recommended
as
an
alternative
to
eliminating
the
notification
requirement
altogether
is
to
exclude
them
from
title
V
and
other
air
pollution
permit
requirements.
One
commenter
(
599)
requested
that
the
notification
requirement
be
removed
for
units
which
meet
the
exceptions
noted
under
§
63.6590(
b).
The
commenter
indicated
that
the
initial
notification
requirement
in
itself
is
not
burdensome,
but
it
will
in
many
States
require
the
commenter
to
file
revisions
or
amendments
to
their
title
V
permits.
This
will
not
only
require
resources
of
the
regulated
community
but
also
add
burdens
upon
the
States
to
process
these
revisions
or
amendments.

Response:
The
EPA
does
not
agree
with
the
commenters
and
does
not
think
this
will
be
a
burden.
An
initial
notification
is
not
a
time
consuming
activity.
Initial
notifications
will
indicate
75
to
EPA
the
number
of
facilities
and
engines,
and
the
type
of
engines
that
will
be
subject
to
the
final
rule,
and
will
clarify
the
categorization
of
such
engines.
The
EPA
is
not
requiring
existing
emergency
engines
to
submit
an
initial
notification.
The
EPA
is
not
requiring
existing
limited
use
engines
to
submit
an
initial
notification
either.
In
addition,
existing
engines
that
combust
landfill
gas
or
digester
gas
equivalent
to
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis
are
also
not
required
to
submit
an
initial
notification.
All
new
emergency
engines,
new
limited
use
engines,
and
new
engines
that
combust
landfill
or
digester
gas
equivalent
to
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis
must
submit
an
initial
notification.

V.
C.
2
Comment:
Four
commenters
(
542,
554,
556,
587)
asked
that
EPA
clarify
that
no
notice
is
required
for
existing
2SLB
and
4SLB
engines.
Three
commenters
(
542,
554,
587)
requested
that
EPA
clarify
that
for
emergency
generators,
notice
is
only
required
for
existing
4SRB
engines
greater
than
500
HP.
One
commenter
(
544)
recommended
clarifying
§
63.6590(
b)
so
that
it
will
not
be
interpreted
that
emergency/
limited
use
and
digester/
landfill
gas
units
500
HP
or
less
need
to
submit
initial
notification.

Response:
The
EPA
feels
it
has
written
the
final
rule
in
a
manner
that
is
clear.
No
notice
is
required
for
existing
2SLB,
existing
4SLB,
or
existing
CI
engines.
No
notice
is
required
for
existing
emergency,
existing
limited
use,
or
existing
engines
that
combust
landfill
gas
or
digester
gas
equivalent
to
10
percent
or
more
of
the
gross
heat
input
on
an
annual
basis.
All
new
engines
must
submit
an
initial
notification.
Also,
EPA
has
clarified
the
final
rule
to
indicate
that
engines
500
HP
or
less
are
not
affected
units
under
the
final
rule,
and
therefore
would
not
be
subject
to
any
notification
requirements.

D.
Startup,
Shutdown,
Malfunction
Plan
V.
D.
1
Comment:
One
commenter
(
551)
argued
that
the
requirement
for
an
immediate
startup,
shutdown,
and
malfunction
(
SSM)
report
should
indicate
that
this
is
required
only
when
the
actions
addressing
the
malfunction
were
inconsistent
with
the
SSMP.

Two
commenters
(
544,
564)
stated
that
EPA
should
eliminate
the
immediate
SSM
report
indicated
in
Table
7,
item
2,
of
the
proposed
rule.
Commenter
564
further
noted
that
any
reporting
requirements
should
be
consistent
with
the
General
Provisions
and
the
December
2002
proposal
relating
to
reporting
malfunctions
only
versus
startups
and
shutdowns.

Two
commenters
(
569,
581)
recommended
eliminating
the
requirement
for
an
immediate
SSM
plan
(
SSMP)
in
Table
7
of
the
proposed
rule.

Response:
The
EPA
agrees
that
immediate
SSMP
reports
are
unnecessary
and
have
the
potential
of
becoming
a
burdensome
activity
for
sources
with
frequent
startups
and
shutdowns.
The
EPA
has
specified
in
the
final
rule
that
an
immediate
SSMP
report
is
only
required
when
actions
addressing
the
startup,
shutdown,
or
malfunction
were
inconsistent
with
the
SSMP.

V.
D.
2
Comment:
One
commenter
(
569)
was
of
the
opinion
that
the
reporting
of
startups
76
and
shutdowns
in
compliance
reports
serves
no
useful
purpose.

Response:
The
EPA
disagrees
that
the
reporting
of
startups
and
shutdowns
in
compliance
reports
does
not
serve
a
useful
purpose.
This
type
of
reporting
provides
information
to
the
enforcing
agency
about
the
typical
operation
of
the
engine
and
therefore
will
help
the
enforcing
agency
better
understand
and
identify
problems.

V.
D.
3
Comment:
Sixteen
commenters
(
526,
536,
540,
544,
549,
551,
558,
559,
564,
565,
569,
570,
578,
579,
581,
585)
said
that
Table
8,
item
16
of
the
proposed
rule
indicates
that
no
SSMP
is
required.
However,
§
63.6640(
c)
requires
a
source
to
be
operated
in
accordance
with
the
SSMP.
Seven
commenters
(
526,
540,
564,
565,
569,
578,
579)
stated
that
RICE
units
do
not
need
SSMP.
One
commenter
(
540)
said
that
§
63.6650(
c)(
4)
should
be
removed
and
Table
8,
item
81
of
the
proposed
rule
should
be
changed
to
"
No."
Commenter
544
indicated
that
there
is
no
justification
for
this
high
level
of
reporting.
This
commenter
further
stated
that
for
consistency,
item
15
in
Table
8
of
the
proposed
rule
should
be
modified
to
include
an
explanation
that
compliance
with
§
63.6(
e)(
1)­(
2)
is
required
with
the
exception
of
a
SSMP,
which
is
not
required.
Three
commenters
(
549,
551,
559)
requested
that
the
requirement
for
SSMPs
be
limited
to
malfunctions.
One
commenter
(
559)
stated
that
Table
8,
item
16
of
the
proposed
rule
should
be
changed
to
require
compliance
with
the
malfunction
components
of
§
63.6(
e)(
3)
(
requiring
SSMPs).
Three
commenters
(
536,
558,
585)
suggested
clarifying
that
SSMPs
are
only
required
for
control
devices.
(
If
EPA
does
include
SSMP
requirements.)
Commenter
564
said
that
the
regulatory
text
should
be
corrected.
The
commenter
stated
that
consistent
with
the
requirement
for
no
SSM
plan,
the
Agency
will
need
to
make
corrections
to
several
references
to
the
General
Provisions
in
Table
8
of
the
proposed
rule.
These
include:

§
63.6(
e)(
1)(
i):
Clarify
that
emission
limitations
and
operating
limitations
do
not
apply
during
SSM
periods.

§
63.6(
e)(
1)(
ii):
Eliminate
reference
to
the
SSM
plan.

§
63.8(
c)(
iii):
Eliminate
the
requirement
for
a
CMS
SSM
plan.

§
63.10(
b)(
2)(
iv
&
v):
Eliminate
reference
to
SSM
plan.

§
63.10(
c)(
15):
Eliminate
reference
to
SSM
plan.

§
63.10(
d)(
5):
Eliminate
this
requirement.

Response:
The
EPA
agrees
that
there
were
conflicting
SSM
requirements
in
the
proposal.
However,
the
EPA
does
not
agree
with
the
commenters
that
a
SSMP
is
not
necessary.
Developing
and
implementing
a
written
SSMP
will
ensure
that
owners
and
operators
maintain
and
operate
their
engines
as
properly
as
possible
during
startup,
shutdown,
and
malfunction
and
will
also
ensure
that
owners
and
operators
minimize
their
emissions
during
these
times.
The
EPA
does
not
feel
it
is
appropriate
to
require
SSMPs
only
for
control
devices;
sources
must
operate
77
their
engines
including
associated
air
pollution
control
and
monitoring
equipment
according
to
the
SSMP.
This
is
consistent
with
the
requirements
of
the
General
Provisions.
The
EPA
has
clarified
in
the
final
rule
that
a
SSMP
is
required.
The
EPA
has
therefore
not
changed
§
63.6650(
c)(
4),
the
applicability
of
§
63.10(
d)(
5)
in
Table
8
of
the
final
rule,
or
made
the
change
to
§
63.6(
e)(
1)­(
2)
that
was
recommended
by
a
commenter.
The
EPA
has
noted
that
§
63.6(
e)(
1)
is
a
reserved
paragraph.
The
requirement
for
SSMPs
are
not
limited
to
malfunctions
but
must
also
include
startups
and
shutdowns,
which
will
ensure
that
sources
operate
their
engines
as
properly
as
possible
during
these
times
as
well
as
during
malfunctions.

V.
D.
4
Comment:
One
commenter
(
555)
observed
that
periods
of
startup
and
shutdown
are
not
defined
in
the
rule.
The
commenter
recommended
that
a
maximum
period
for
startup
and
shutdowns
be
defined
in
the
rule
with
the
possibility
of
the
Administrator
allowing
a
longer
period
on
a
case­
by­
case
basis.

Response:
The
EPA
assumes
that
this
comment
was
made
because
Table
8
in
the
proposed
rule
indicated
that
a
SSMP
was
not
required.
However,
EPA
has
specified
in
the
final
rule
that
a
SSMP
is
required
to
be
developed
and
maintained
and
therefore
feels
that
this
satisfies
the
commenter's
concern.

V.
D.
5
Comment:
One
commenter
(
549)
suggested
that
EPA
include
a
discussion
of
the
operating
profile
for
engines
in
section
III.
H
of
the
preamble
and
state
that
reporting
for
startup
and
shutdown
is
not
warranted
and
that
operational
anomalies
will
be
addressed
by
malfunction
reporting.

Response:
The
EPA
does
not
feel
it
is
necessary
to
include
a
discussion
of
the
operating
profile
for
engines
in
the
preamble.
Sources
are
required
to
report
every
startup
and
shutdown
in
the
compliance
reports.
Sources
are
required
to
report
when
actions
addressing
the
startup,
shutdown,
or
malfunction
were
inconsistent
with
the
SSMP.

V.
D.
6
Comment:
One
commenter
(
569)
expressed
support
for
EPA's
position
that
deviations
that
occur
during
periods
of
SSMs
are
not
violations.

Response:
The
EPA
has
noted
the
commenter's
support.

E.
Reports
V.
E.
1
Comment:
Two
commenters
(
570,
577)
requested
annual
compliance
reports
instead
of
the
requirement
of
semiannual
reporting
of
compliance
reports
in
§
63.6650(
3).
Commenter
570
asked
that
the
language
in
this
paragraph
be
modified
to
allow
the
flexibility
for
annual
compliance
reports
in
order
to
make
the
final
rule
consistent
with
other
MACT
standards.
The
commenter
noted
that
they
are
seeing
in
the
various
State
and
Federal
regulations
the
requirements
for
monthly,
quarterly,
semiannual,
and
annual
reports,
and
keeping
track
of
these
is
becoming
quite
difficult.
Commenter
577
stated
that
this
will
create
an
unnecessary
paperwork
burden
for
both
the
regulated
community
as
well
as
for
the
regulatory
agencies.
A
more
78
reasonable
approach
would
be
to
require
an
annual
compliance
report
timed
concurrently
with
the
state
EPA's
typical
emissions
reporting
requirement.

Response:
The
EPA
disagrees
that
semiannual
compliance
reports
are
a
burden.
The
EPA
feels
that
the
submittal
of
semiannual
reports
will
assist
in
identifying
problem
areas
within
a
reasonable
period
of
time.
The
requirement
for
semiannual
compliance
reporting
is
not
inconsistent
with
previous
MACT
standards.
Several
MACT
standards
require
compliance
reports
to
be
prepared
and
submitted
semiannually.
Enforcing
agencies
have
been
requiring
semiannual
compliance
reports
for
a
long
time,
and
this
has
worked
well
and
has
helped
EPA
enforce
rules
appropriately.
The
EPA
feels
the
submittal
of
semiannual
compliance
reports
is
appropriate
for
stationary
RICE
complying
with
the
final
rule.

F.
Other
V.
F.
1
Comment:
One
commenter
(
569)
stated
that
readily
available
electronic
records
do
not
have
to
be
stored
on­
site.
In
§
63.6660(
c),
the
proposed
RICE
MACT
requires
that
records
be
kept
on­
site
for
the
first
2
years
following
the
date
of
each
occurrence,
measurement,
maintenance,
corrective
action,
report
or
record.
This
requirement
does
not
recognize
the
trend
toward
computerization
of
monitoring
records.
Many
sites
are
making
an
intentional
effort
to
move
away
from
paper
records
of
air
compliance
critical
data
whenever
the
opportunity
presents
itself.
These
electronic
records
reside
on
hardware
referred
to
as
servers.
For
a
variety
of
reasons,
these
servers
are
not
always
located
at
the
major
source
that
would
be
affected
by
the
RICE
MACT.
There
are
cases
at
companies
where
the
server
for
an
affected
source
is
not
located
in
the
same
State
as
the
affected
source.
The
concept
of
"
readily
accessible"
should
be
more
important,
relative
to
current
records,
than
the
need
for
them
to
be
on­
site
at
the
major
source.
The
commenter
urged
EPA
to
recognize
the
trend
to
electronic
record
keeping
by
changing
§
63.6660(
c)
to
read
as
follows:

"(
c)
Each
record
must
be
readily
accessible
in
hard
copy
or
electronic
form
on­
site
for
at
least
2
years
after
the
date
of
each
occurrence,
measurement,
maintenance,
corrective
action,
report
or
record
according
to
§
63.10(
b)(
1).
You
may
keep
the
records
off­
site
for
the
remaining
3
years."

Response:
The
EPA
agrees
with
the
commenter
and
feels
that
records
that
can
be
accessed
on­
site
by
a
computer
are
valid
and
should
be
considered
on­
site
records.
The
EPA's
understanding
of
the
General
Provisions
is
that
it
allows
the
interpretation
that
records
that
can
be
accessed
on­
site
are
acceptable.
In
any
case,
EPA
has
written
§
63.6660(
c)
in
the
final
rule
according
to
the
commenter's
suggestion.

V.
F.
2
Comment:
One
commenter
(
549)
recommended
that
EPA
resolve
the
inconsistency
in
Table
5
of
the
proposed
rule
to
reflect
the
same
averaging
period.
In
Table
5
1.
a.
i
and
ii
of
the
proposed
rule,
the
averaging
period
is
inconsistent
with
other
references
in
the
table.
The
averaging
period
should
be
made
the
same
throughout
the
table.
79
Response:
The
commenter
does
not
explain
what
inconsistencies
it
is
referring
to
in
Table
5
of
the
proposed
rule.
The
EPA
disagrees
with
the
commenter
and
cannot
see
any
inconsistencies
in
Table
5
of
the
proposed
rule
and
has
therefore
not
made
any
changes.

V.
F.
3
Comment:
Three
commenters
(
549,
565,
567)
stated
that
the
continuous
compliance
requirements
should
be
modified
to
specify
a
24­
hour
block
average
rather
than
a
4­
hour
rolling
average
for
demonstrating
compliance.
Commenter
565
stated
that
the
proposed
boiler
MACT
has
a
1­
calendar
day
(
i.
e.,
24
hour)
averaging
time
for
CO
CEMS
associated
with
the
400
ppmv
CO
emission
limit
for
new/
reconstructed
units.

Response:
The
EPA
feels
that
the
4­
hour
averaging
period
is
justified
and
is
appropriate
for
engines
and
has
therefore
retained
this
averaging
period
in
the
final
rule.
Gas
and
diesel­
fired
units
have
less
variability
in
the
feed
stream
than
other
sources
such
as
coal­
fired
boilers;
therefore
a
shorter
averaging
time
is
appropriate
for
those
units.
The
4­
hour
averaging
period
is
similar
to
the
typical
time
associated
with
performance
tests
(
3­
4
hours).
Four
hours
is
also
a
sufficient
time
period
to
describe
the
typical
operation
of
an
engine
and
is
more
representative
than
24
hours.
The
EPA
feels
that
a
24­
hour
averaging
period
is
not
appropriate
for
engines.
Any
variation
in
emissions
that
might
be
expected
to
occur
may
not
be
caught
by
the
24­
hour
averaging
period.
For
these
reasons,
EPA
feels
that
the
4­
hour
averaging
period
is
appropriate.

V.
F.
4
Comment:
One
commenter
(
570)
remarked
that
EPA
should
be
more
explicit
in
§
63.6620(
e)(
2)
and
state
what
the
equivalent
CO
2
percentage
should
be.
The
EPA
should
also
specify
how
this
normalization
should
be
done,
e.
g.,
by
referencing
40
CFR
60,
appendix
A,
Reference
Method
20.

Response:
The
EPA
agrees
with
the
commenter
and
has
provided
more
explicit
instructions
from
40
CFR
60,
appendix
A,
reference
method
20
in
the
final
rule.
The
equivalent
CO
2
percentage
varies
depending
on
the
fuel.
The
EPA
can
therefore
not
specify
what
the
equivalent
percentage
should
be.
The
CO
2
concentration
at
stoichiometric
conditions
must
first
be
calculated.
Then
the
CO
2
equivalent
is
calculated
as
5.9/
20.9
times
that
value
(
CO
2
ultimate).
V.
F.
5
Comment:
One
commenter
(
545)
stated
that
the
Naval
Air
Weapons
Station
(
NAWS)
China
Lake
has
several
water
pumps
powered
by
stationary
RICE
at
its
rocket
motor
testing
complex.
These
pumps
provide
deluge
and/
or
quench
water
during
certain
types
of
rocket
motor
testing.
They
operate
less
than
one
hour
during
any
one
test
and
only
one
test
involving
the
deluge
pumps
can
be
performed
on
any
one
day.
During
a
typical
rocket
motor
test,
these
pumps
operate
at
"
full
load,"
pumping
water
for
only
5
to
15
minutes
and
idle
during
startup
and
shutdown
periods
the
remainder
of
the
hour.
The
stationary
RICE
units
powering
the
pumps
should
be
considered
limited
use
even
though
they
may
be
used
more
than
50
hours
per
year
depending
on
the
rocket
motor
test
schedule.
In
addition,
most
of
these
operating
hours
take
place
during
stationary
RICE
startup
or
shutdown
periods
when
the
proposed
rule's
emission
limitations
and
operating
limitations
do
not
apply.

The
commenter
also
stated
that
the
proposed
rule
requires
owners
and
operators
complying
with
the
CO
emission
reduction
requirement
by
using
an
oxidation
catalyst
to
install,
operate,
and
80
maintain
a
CEMS.
The
CEMS
must
complete
at
least
one
cycle
of
operation
every
15
minutes
and
must
have
at
least
two
data
points
for
successive
15­
minute
periods
within
the
same
hour
for
a
"
valid"
hour
of
monitoring
data.
For
the
NAWS
China
Lake
deluge
pump
stationary
RICE,
the
time
between
startup
and
shutdown
periods
would
barely,
if
at
all,
compromise
one
15­
minute
monitoring
cycle
making
it
difficult,
if
not
impossible,
to
obtain
two
data
points
during
the
same
hour.
Therefore,
if
CEMS
were
installed,
operated
and
maintained
on
these
stationary
RICE,
NAWS
China
Lake
may
not
be
able
to
obtain
a
"
valid"
1
hour
of
CEMS
data
as
required
by
§
63.6625
of
the
proposed
rule.

In
addition,
the
commenter
stated
that
the
proposed
rule
requires
certain
owners
and
operators
to
install,
operate,
and
maintain
a
CPMS
according
to
the
requirements
of
40
CFR
section
63.8
of
the
General
Provisions.
Although
the
proposed
rule
does
not
specify
the
required
frequency
of
CPMS
monitoring
cycles,
it
is
conceivable
that
the
NAWS
China
Lake
deluge
pump
stationary
RICE
operating
profile
would
also
make
CPMS
use
impractical.
Additionally,
Table
6
of
the
proposed
rule
requires
CEMS
and
CPMS
monitoring
data
to
be
reduced
to
4­
hour
rolling
averages.
The
commenter
stated
that
it
operates
its
deluge
pumps
less
than
1
hour
in
a
day,
it
may
not
be
possible
to
obtain
a
"
valid"
4­
hour
rolling
average
as
required
by
the
proposed
rule.

Response:
Since
proposal
EPA
has
increased
the
number
of
hours
allowed
for
limited
use
operation.
In
the
final
rule,
EPA
has
defined
limited
use
as
those
units
that
operate
less
than
100
hours
per
year.
The
commenter
stated
that
the
stationary
RICE
it
is
concerned
with
operate
less
than
1
hour
during
any
one
test
and
only
one
test
involving
the
deluge
pumps
can
be
operated
on
any
one
day.
In
addition,
the
commenter
also
stated
that
during
a
typical
rocket
motor
test,
these
units
pump
water
for
only
5
to
15
minutes
and
are
idle
during
startup
and
shutdown
periods
the
remainder
of
the
hour.
As
the
commenter
mentioned,
emissions
limitations
and
operating
limitations
do
not
apply
during
startup
and
shutdown
periods.
The
stationary
RICE
then
would
only
have
to
meet
the
emission
limitations
and
operating
limitations
during
the
times
it
is
actually
operating;
a
maximum
of
15
minutes
per
day
or
approximately
90
hours
if
operated
every
day
for
1
year.
However,
since
EPA
has
increased
the
number
of
hours
allowed
for
limited
use
operation,
the
stationary
RICE
the
commenter
mentioned
would
be
considered
limited
use
units,
and
would
not
have
to
meet
the
emission
limitations
or
operating
limitations
of
the
final
rule.

The
final
rule
does
not
specify
that
CO
CEMS
are
required.
The
final
rule
specifies
that
sources
that
meet
the
CO
percent
reduction
emission
limitation
can
use
portable
CO
monitors
to
demonstrate
compliance
in
addition
to
maintaining
operating
limitations.
Therefore,
the
commenters
concerns
regarding
CO
CEMS
are
moot.
Sources
have
the
option
to
install
and
operate
a
CO
CEMS
to
demonstrate
compliance
with
the
CO
percent
reduction
requirement,
but
it
is
not
a
requirement.

V.
F.
6
Comment:
One
commenter
(
549)
stated
that
EPA
should
resolve
the
inconsistency
in
Table
4
of
the
proposed
rule
regarding
choice
of
test
methods.
Table
4,
2.
a.
ii(
1)
of
the
proposed
rule
should
read
method
3A
or
3B
to
be
consistent
with
Table
4,
3.
a.
ii(
1).

Response:
The
EPA
agrees
with
the
commenter
and
has
included
methods
3A
and
3B
in
81
Table
4,
2.
a.
ii(
1)
of
the
final
rule.

VI.
TESTING
A.
Test
Methods
1.
CARB
430
VI.
A.
1.1
Comment:
Five
commenters
(
526,
542,
554,
556,
587)
pointed
out
that
there
is
a
50
ppm
NOx
limit
advisory
with
the
use
of
CARB
Method
430.
The
commenters
asked
EPA
to
follow
the
direction
of
the
CARB
advisory.
Commenter
587
added
that
due
to
concerns
about
matrix
interferences
with
CARB
Method
430,
as
expressed
in
an
advisory
released
by
CARB,
the
commenter
stated
that
it
is
inappropriate
to
include
CARB
Method
430
as
a
candidate
method
until
its
governing
agency
has
more
thoroughly
researched
method
deficiencies
and
revised
the
method
or
rescinded
the
advisory.

Response:
The
EPA
agrees
that
CARB
Method
430
use
should
not
be
cited
in
the
final
rule.
Therefore,
the
EPA
has
not
included
CARB
Method
430
as
a
test
method
in
the
final
rule.

2.
FTIR
VI.
A.
2.1
Comment:
One
commenter
(
530)
expressed
that
surrogate
measurements
have
their
usefulness,
however,
initial
emission
measurements
and
periodic
(
recommended
3
years)
measurements
should
be
conducted
with
FTIR.
The
commenter
recommended
a
combination
of
periodic
FTIR
measurements
and
frequent
surrogate
measurements.
Current
cost
of
FTIR
measurements
now
is
extremely
competitive
to
that
of
inferior
methods.

Response:
The
EPA
agrees
with
the
commenter
that
FTIR
is
the
most
accurate
and
precise
method
currently
available
to
test
for
formaldehyde
emissions
from
the
stationary
engine
exhaust.
The
EPA
has
included
FTIR
as
an
acceptable
test
method
in
the
final
rule.

V1.
A.
2.2
Comment:
One
commenter
(
587)
stated
that
it
is
imperative
that
multiple
test
methods
and
technological
approaches
be
available
for
formaldehyde
measurement
from
engines.
The
commenter
stated
that
FTIR
methods
address
this
need.

Response:
The
EPA
agrees
with
the
commenter
that
it
is
preferable
that
as
many
valid
test
methods
and
technological
approaches
be
allowed
and
therefore
conducted
a
search,
as
intended
by
the
National
Technology
Transfer
and
Advancement
Act,
in
an
attempt
to
identify
available
national
consensus
based
methods
to
test
for
formaldehyde
from
stationary
engines.
The
EPA
identified
ASTM
D6348­
03
as
a
potential
national
consensus
based
method
in
addition
to
Method
320
and
EPA
Method
323.
Upon
review,
EPA
approved
this
method
as
an
alternative
to
Method
320
for
formaldehyde
measurement
provided
in
ASTM
D6348­
03,
Annex
5
(
Analyte
Spiking
Technique),
percent
R
must
be
greater
than
or
equal
to
70
and
less
than
or
equal
to
130.
82
VI.
A.
2.3
Comment:
One
commenter
(
587)
stated
that
it
does
not
have
specific
recommendations
for
revisions
to
EPA
Method
320
at
this
time,
however,
the
commenter
has
reservations
about
the
EPA
statement
in
the
preamble
that
indicates
that
the
method
will
be
revised,
"
...
to
ensure
it
can
be
used
to
accurately
measure
formaldehyde
concentrations
as
low
as
8
ppbvd
in
the
exhaust
from
a
stationary
RICE."
Based
on
the
commenter's
experience
with
extractive
FTIR
testing,
as
well
as
comments
from
leading
experts
on
extractive
FTIR
measurement
from
combustion
sources,
the
commenter
did
not
believe
that
this
level
of
detection
is
achievable
in
practice.
It
is
premature
for
EPA
to
make
a
statement
in
the
preamble
regarding
such
a
low
detection
limit
for
FTIR.
The
commenter
is
not
aware
of
any
evidence
to
support
such
a
low
detection
limit
for
FTIR.
In
considering
the
revisions
to
Method
320,
it
is
imperative
that
EPA
achieve
scientific
consensus
among
experts
in
the
field,
as
was
done
for
the
development
and
approval
of
the
ASTM
method.
In
addition,
EPA
should
consider
that
the
test
method
must
be
achievable
in
practice
by
professionals
with
a
level
of
expertise,
and
a
required
capital
investment,
commensurate
with
the
current
standards
in
the
emission
source
test
community.

Response:
The
FTIR
technology
is
rapidly
improving
and
EPA
feels
this
trend
will
continue.
Method
320
is
a
self­
validated
performance­
based
method
that
will
allow
the
use
of
this
improved
technology
and
not
suppress
it.
Current
state­
of­
the­
art
methodology
can
identify
and
measure
formaldehyde
concentrations
well
below
the
current
emission
limitation
with
a
path
length
of
10
meters
or
less.
The
method
has
a
pretest
preparation
and
evaluation
procedure
to
determine
the
optimum
sampling
system
configuration
for
measuring
formaldehyde
customized
to
the
technology
the
tester
is
using.
Detection
levels
in
FTIRs
are
limited
either
by
electronic
noise
or
by
interference
from
other
gases.
Measurement
of
formaldehyde
has
very
little
interference
so
one
can
get
close
to
the
noise
limited
detection
with
state­
of­
the­
art
technology.
The
cost
to
conduct
a
test
is
not
the
primary
concern
of
EPA,
however,
the
cost
appears
to
be
decreasing,
as
would
be
expected
in
the
U.
S.
free
market
system.
The
expected
decrease
in
cost
is
also
supported
by
a
previous
commenter
who
indicated
that
current
cost
of
FTIR
measurements
are
at
this
time
extremely
competitive
to
that
of
other
methods.
Water
vapor
(
and
other)
interferences
are
taken
into
consideration
in
the
pretest
planning.
Industry
is
generally
in
agreement
that
Method
320
is
the
most
accurate
and
reliable
test
method
currently
available
to
test
for
formaldehyde
emissions
from
the
stationary
engine
exhaust.
As
such,
EPA
feels
that
it
is
appropriate
to
allow
FTIR
for
the
purpose
of
determining
compliance
with
the
formaldehyde
emission
limitation.

3.
ASTM
D6348
VI.
A.
3.1
Comment:
Five
commenters
(
542,
549,
554,
556,
587)
stated
that
EPA
should
allow
ASTM
Method
D6348
as
equivalent
to
Method
320.
Commenter
587
added
that
the
Method
is
self­
validating
and
includes
clarity
that
the
commenter
stated
will
provide
better
consistency
and
reduce
the
likelihood
of
errors
as
FTIR
becomes
more
widely
implemented
by
the
source
test
community.
The
ASTM
method
was
developed
and
approved
following
a
refereed
process
and
considering
the
input
and
review
of
leading
experts
in
the
field.
83
Response:
The
ASTM
made
a
few
changes
to
ASTM
D6348­
98.
These
were
successfully
balloted.
The
EPA
accepts
the
new
version
as
an
alternative
to
Method
320,
40
CFR
part
63,
appendix
A
for
formaldehyde
measurement
provided
in
ASTM
D6348­
03,
Annex
A5
(
Analyte
Spiking
Technique),
the
percent
R
must
be
greater
than
or
equal
to
70
and
less
than
or
equal
to
130.

4.
EPA
Method
323
VI.
A.
4.1
Comment:
Four
commenters
(
542,
554,
556,
587)
recommended
that
EPA
include
proposed
EPA
Method
323.
Commenter
587
felt
that
it
is
imperative
that
multiple
test
methods
and
technological
approaches
be
available
for
formaldehyde
measurement
from
engines.
The
EPA
Method
323
addresses
this
need
and
appears
to
offer
a
reasonable
alternative
to
FTIR
for
formaldehyde
testing
of
engines.
The
method
detection
limits
are
within
the
range
necessary
to
demonstrate
compliance
with
a
formaldehyde
based
limit.
This
method
was
investigated
and
developed
by
the
Gas
Technology
Institute
(
GTI)
as
a
low­
cost
alternative
for
engine
formaldehyde
measurement
and
has
been
validated
for
application
to
internal
combustion
engines
in
research
conducted
by
GTI.

One
commenter
(
605)
said
that
this
method
has
the
advantage
of
actually
having
been
fieldvalidated
at
the
required
concentration.
Furthermore,
it
is
simpler
and
less
costly
than
the
other
methods.
It
is
the
commenter's
experience
that
with
a
similar
chilled­
impinger
method
for
VOC
(
Method
25.3),
they
found
it
was
critical
to
maintain
near­
ice­
water
temperatures
in
order
to
achieve
100
percent
capture.
The
method
might
be
modified
by
adding
a
final
impinger
and
having
that
analyzed
separately
for
breakthrough.
Sulfur
dioxide
is
listed
as
an
interference,
possibly
because
of
its
ability
to
bond
with
aldehydes.
This
bond
is
broken
under
acidic
conditions.
If
this
is
found
to
be
a
problem,
perhaps
the
sample
can
be
acidified
more
to
break
up
any
complexes.

Response:
The
EPA
agrees
with
the
commenter
and
has
included
EPA
Method
323
as
an
optional
method
for
natural
gas­
fired
units
in
the
final
rule.
The
EPA
plans
to
develop
a
FAQ
sheet
for
EPA
Method
323.
The
EPA
may
include
the
commenter's
suggestion
for
analyzing
for
breakthrough
with
another
impinger
and
a
caution
to
check
the
impinger
exhaust
temperature
when
assessing
the
data
quality.

5.
EPA
SW­
846
Method
0011
VI.
A.
5.1
Comment:
One
commenter
(
587)
expressed
the
view
that
since
EPA
SW­
846
Method
0011
uses
a
similar
analytical
approach
as
CARB
Method
430,
has
not
been
validated
for
application
to
engines,
and
has
quality
assurance
requirements
considered
less
thorough
than
CARB
Method
430,
it
should
be
excluded
from
the
list
of
acceptable
methods.

Response:
The
EPA
agrees
with
the
commenter
that
this
method
should
not
be
specified
as
an
acceptable
method
for
this
application.
This
method
has
not
been
included
in
the
final
rule.
84
6.
Other
VI.
A.
6.1
Comment:
One
commenter
(
500)
indicated
that
two
ASTM
standards
have
been
updated.
E
337­
84
is
now
E
337­
02
and
D
3154­
91
is
now
D
3154­
00.

Response:
The
EPA
has
noted
the
comment
and
has
made
the
appropriate
updates
in
the
final
rule.

VI.
A.
6.2
Comment:
One
commenter
(
552)
asserted
that
EPA
must
carefully
determine
the
appropriateness
of
any
specified
test
procedure.

Response:
The
EPA
has
carefully
determined
the
appropriateness
of
the
specified
test
procedures
in
the
final
rule.
The
EPA
has
removed
CARB
Method
430
and
EPA
SW­
846
Method
0011
from
the
final
rule
as
discussed
in
previous
responses
to
comments.
The
EPA
has
specified
in
the
final
rule
that
FTIR,
EPA
Method
323,
or
ASTM
D6348­
03
should
be
used
to
demonstrate
compliance
with
formaldehyde
emission
limitations.

VI.
A.
6.3
Comment:
One
commenter
(
530)
urged
EPA
to
realize
that
the
high
level
of
variations
and
biases
observed
with
outdated,
indirect,
and
inferior
methods
(
CARB
430,
EPA
SW­
846
Method
0011,
EPA
323)
may
lead
to
future
problems.
FTIR
has
been
demonstrated
to
be
the
most
accurate,
precise
and
cost
effective
method
known
for
formaldehyde
measurements.
The
commenter
added
that
it
is
wrong
to
propose
methods
that
indirectly
measure
formaldehyde
via
limited
physical
processes
and
chemical
reactions
when
a
direct
method
exists.
When
possible
the
actual
pollutant
must
be
measured
in
order
to
determine
how
much
of
it
is
present
so
it
can
be
regulated.
FTIR
also
has
the
ability
to
allow
optimization
of
emissions
in
real­
time
which
insures
the
lowest
emissions
operation
of
a
given
process.

Response:
The
EPA
agrees
with
the
commenter
that
properly
used,
FTIR
results
is
the
more
certain
measurement
of
formaldehyde.
The
EPA
has
not
specified
CARB
430
or
SW
846
0011
in
the
final
rule
but
has
retained
EPA
Method
323
as
an
alternative
since
EPA
has
reviewed
a
valid
(
Method
301)
GTI
report
that
indicates
acceptable
precision
of
about
11
percent
and
recoveries
of
formaldehyde
of
about
110
percent
on
a
2SLB
engine.
In
addition,
the
National
Council
on
Air
and
Stream
Improvement
validated
a
similar
method
for
several
wood
product
sources
and
several
pulp
and
paper
sources
with
similar
or
better
precision
and
acceptable
recoveries.
The
paired
data
submitted
by
commenter
588
does
not
exhibit
this
same
level
of
precision
expected
from
this
method.
The
comparison
of
paired
data
sets
appear
almost
random.
Consequently,
EPA
suspects,
but
cannot
find
proof
in
the
report
that
there
is
some
sampling
or
analysis
anomaly
in
this
data
set.
Likewise,
given
the
lack
of
precision,
EPA
hesitates
to
reject
EPA
Method
323
based
on
comparing
the
FTIR
results
with
the
mean
of
these
dissimilar
data
when
EPA
has
historical
data
successfully
validating
similar
methods
on
a
wide
variety
of
wood
product
and
pulp
and
paper
sources.

VI.
A.
6.4
Comment:
One
commenter
(
570)
observed
that
Table
4
of
the
proposed
rule
does
not
appear
to
allow
the
option
of
EPA
Reference
Method
3.
The
commenter
questioned
85
why
this
method
is
not
included
and
asked
that
the
rule
be
written
to
allow
for
this
option.

Response:
The
EPA
agrees
that
Method
3
can
be
used
and
made
this
change
to
the
final
rule.

VI.
A.
6.5
Comment:
One
commenter
(
588)
indicated
that
Wyoming
has
been
requiring
formaldehyde
testing
to
gather
data.
Three
test
methods
(
FTIR,
EPA
SW­
846
Method
0011,
and
Acetyl
Acetone
(
Celanese))
have
been
used.
The
commenter's
review
of
the
Celanese
method
or
proposed
EPA
Method
323
is
that
it
provides
inconsistent
results.
The
commenter
assumed
that
the
Celanese
method
went
through
a
validation
process.
If
FTIR
is
the
validating
method,
the
commenter
questioned
whether
there
is
an
acceptable
correlation
between
the
two
methods.
The
commenter
would
like
to
see
the
validation
data.
The
commenter
has
included
one
test
report
which
compares
FTIR
to
Celanese
methods.

Response:
The
EPA
agrees
with
the
commenter
that
the
data
indicate
significant
differences
in
concentration
from
run
to
run.
However
since
in
four
sets
the
Celanese
method
was
larger
than
the
FTIR
while
the
other
three
times
it
was
smaller,
EPA
hesitates
to
draw
any
other
conclusions.
The
validation
data
have
been
added
to
the
docket.

VI.
A.
6.6
Comment:
One
commenter
(
605)
contended
that
EPA
Method
320
(
Extractive
FTIR),
CARB
Method
430
(
in­
field
DNPH
derivatization),
and
EPA
SW­
846
Method
0011
(
similar
to
CARB
430)
should
not
be
cited
in
the
proposed
rule
because
they
have
not
been
validated
for
the
sources
and
concentrations
in
questions.

Response:
Although
EPA
disagrees
with
the
commenters
rationale,
EPA
agrees
that
CARB
Method
430
and
EPA
SW­
846
Method
0011
should
not
be
cited
in
the
final
rule.
Method
320
is
a
self­
validating
performance
based
method,
that
is
capable
of
being
successfully
implemented
and
is
cited.

B.
Performance
Tests
1.
Portable
CO
Monitor
VI.
B.
1.1
Comment:
Two
commenters
(
526,
579)
stated
that
the
proposed
standards
require
continuous
monitoring
of
pressure
drop
across
the
catalyst,
catalyst
inlet
temperature
and
temperature
rise
across
the
catalyst,
and
setting
operating
limits
of
these
parameters
in
reference
to
the
initial
performance
test
data.
Since
these
operating
parameters
vary
depending
upon
engine
load,
the
initial
performance
test
must
be
mapped
over
the
entire
operating
range
to
establish
the
reference
points.
This
is
especially
true
of
variably
loaded
engines
such
as
those
following
diurnal
sewage
flow
patterns.
Upon
changing
as
little
as
a
single
catalyst
element,
the
same
parameters
must
be
re­
established.
Hence
extremely
comprehensive
and
frequent
source
tests
could
result
from
the
proposed
rule.
The
commenters
stated
that
the
periodic
monitoring
provisions
that
have
been
relatively
recently
agreed
to
by
EPA
and
States
as
part
of
the
title
V
program
should
serve
as
the
continuous
compliance
criteria
of
the
proposed
rule.
The
use
of
simple
hand­
held
CO
86
monitors
using
approved
EPA
protocols
with
results
reported
through
the
title
V
process
are
adequate
and
should
be
offered
as
an
option
to
operators.

Response:
The
EPA
disagrees
and
feels
that
some
type
of
continuous
monitoring
is
necessary
to
demonstrate
continuous
compliance.
The
EPA
has
removed
the
requirement
to
monitor
and
maintain
the
temperature
rise
across
the
catalyst
for
4SRB
engines
using
NSCR.
The
EPA
has
also
adjusted
the
requirement
for
pressure
drop
monitoring
to
only
require
that
the
pressure
drop
across
the
catalyst
be
measured
monthly.
Finally,
EPA
has
removed
the
requirement
to
monitor
and
maintain
the
operating
load
or
fuel
flow
rate
for
sources
complying
with
the
alternative
formaldehyde
concentration
emission
limitation.
The
rationale
for
these
changes
are
discussed
elsewhere
in
this
document.

VI.
B.
1.2
Comment:
Four
commenters
(
542,
554,
556,
587)
stated
that
quarterly
emission
testing
with
CO
portable
units
should
not
be
full
performance
tests.
This
provision
is
burdensome
and
unnecessary.
The
final
rule
should
not
require
that
the
quarterly
emission
tests
be
full
performance
tests
for
the
following
reasons:

(
1)
For
full
performance
tests,
engines
in
load­
following
applications
may
need
to
conduct
emissions
testing
at
multiple
operating
conditions,
in
accordance
with
the
General
Provisions'
requirement
that
performance
tests
be
conducted
for
representative
conditions;

(
2)
Facilities
with
load­
following
operations,
such
as
natural
gas
transmission
and
storage,
may
not
be
able
to
operate
the
engines
over
the
full
range
of
operating
conditions
on
a
quarterly
basis;

(
3)
Full
performance
tests
impose
significant
burden
on
the
owner
or
operator
to
develop
site­
specific
test
plans,
provide
notification
to
the
permitting
authority
60
days
in
advance
of
the
test,
and
submit
the
full
results
within
60
days
of
completion
of
the
testing;
and
(
4)
Review
of
other
MACT
standards
indicates
that
full
performance
tests
are
not
required
more
frequently
than
annually.

Response:
The
EPA
agrees
with
the
commenters
that
requiring
full
performance
tests
quarterly
for
sources
complying
with
CO
reduction
requirement
may
impose
significant
burden
on
the
owner
or
operator
to
develop
site­
specific
test
plans,
provide
notification
to
the
permitting
authority
60
days
in
advance
of
the
test,
and
submit
the
full
results
within
60
days
of
completion
of
the
testing.
The
EPA
now
feels
that
quarterly
testing
for
CO
is
unnecessary
and
inappropriate.
In
the
final
rule,
EPA
has
specified
that
new
2SLB,
new
4SLB,
and
new
CI
engines
complying
with
requirement
to
reduce
CO
emissions
must
conduct
semiannual
performance
tests
for
CO
to
demonstrate
that
the
required
CO
percent
reduction
is
achieved.
Semiannual
performance
testing
for
CO
in
addition
to
monitoring
and
maintaining
operating
parameters
will
ensure,
on
an
ongoing
basis,
that
the
applicable
CO
percent
reduction
requirement
is
being
met.
After
demonstrating
87
compliance
for
two
consecutive
tests,
the
frequency
can
be
reduced
to
annually.
However,
if
an
annual
performance
test
indicates
a
deviation
of
CO
emissions
from
the
CO
reduction
requirement,
you
must
return
to
semiannual
performance
tests.

VI.
B.
1.3
Comment:
One
commenter
(
555)
indicated
that
accurate
measurements
with
portable
emission
monitors
will
be
difficult.
The
portable
the
commenter
is
familiar
with
have
a
full
range
scale
of
0­
500
ppmvd
and
are
not
expected
to
be
accurate
at
10
ppmvd.

Response:
That
particular
instrument
arrangement
would
probably
not
meet
Section
8.3.1
of
ASTM
D
6522­
00
which
requires
the
span
gas
concentration
to
be
chosen
such
that
the
average
stack
gas
reading
for
each
run
is
greater
than
25
percent
of
the
span
gas
concentration
and
also
meet
the
performance
and
design
criteria
of
Section
9
and
10.4.11
as
required
by
Section
7.2
of
the
method.
However,
there
are
portable
monitors
for
sale
that
do
meet
ASTM
D6522­
00
and
our
monitoring
requirements.

2.
Other
VI.
B.
2.1
Comment:
Five
commenters
(
542,
554,
556,
565,
587)
contended
that
additional
performance
tests
should
not
be
required
when
NSCR
or
oxidation
catalysts
are
replaced
with
identical
units.

Response:
The
EPA
disagrees.
Additional
performance
tests
are
required
to
be
performed
even
though
an
emission
control
device
is
replaced
with
an
identical
unit.
The
performance
of
identical
catalysts
can
vary
significantly,
and
it
is
not
guaranteed
that
the
NSCR
or
oxidation
catalyst
will
achieve
the
same
performance
levels.

VI.
B.
2.2
Comment:
One
commenter
(
570)
asked
that
EPA
specify
what
is
meant
by
the
phrase
"
catalyst
elements."
The
commenter
requested
that
EPA
specify
within
the
rule
what
specifically
is
intended
to
trigger
the
requirement
to
conduct
another
performance
test
when
replacing
catalyst
elements.

Response:
Sources
have
to
conduct
another
performance
test
when
the
catalyst
is
replaced.

VI.
B.
2.3
Comment:
One
commenter
(
570)
recommended
that
EPA
eliminate
Table
3
of
the
proposed
rule
and
impose
no
testing
requirements
beyond
the
initial
performance
tests.
One
commenter
(
569)
said
that
the
requirements
for
subsequent
performance
tests
in
Table
3
of
the
proposed
rule
should
be
deleted
from
RICE
equipped
with
catalyst.
Subsequent
performance
tests,
if
any,
should
only
be
required
on
one
RICE
if
a
facility
operates
several
RICE
of
the
same
design.

Response:
The
EPA
disagrees
with
the
commenters.
Subsequent
performance
testing
is
needed
for
units
in
order
to
demonstrate
continuous
compliance
with
the
applicable
emission
limitation
beyond
initial
compliance
demonstration.
Subsequent
performance
tests
must
be
88
conducted
on
each
RICE
even
though
a
facility
operates
several
RICE
of
the
same
design.
The
performance
of
different
stationary
RICE
can
vary
significantly,
even
if
they
are
of
the
same
design.

VI.
B.
2.4
Comment:
One
commenter
(
552)
suggested
that
EPA
include
a
provision
that
performance
tests
be
conducted
within
the
manufacturer's
recommended
operating
range,
excluding
idle.
The
EPA
needs
to
identify
reasonable
engine
operating
and
load
conditions
to
be
used
during
the
initial
and
subsequent
performance
tests.
Tables
5
and
6
of
the
proposed
rule
establishes
requirements
related
to
initial
and
subsequent
performance
tests.
However,
no
direction
is
given
in
the
tables
or
the
text
of
the
proposed
rule
regarding
the
operating
load
to
be
used
during
these
tests.
The
EPA
should
revise
the
tables
and
the
text
of
the
proposed
rule
to
clarify
the
engine
conditions
and
loading
requirements
for
these
tests.
Engine
loads
and
operating
conditions
can
vary
significantly
depending
on
specific
applications
and
changing
demands,
and
it
is
important
that
the
final
rule
does
not
restrict
nor
confine
the
flexibility
needed
to
successfully
operate
these
engines.
The
commenter
therefore
suggested
that
EPA
include
a
provision
that
performance
tests
be
conducted
within
the
manufacturer's
recommended
operating
range.
Since
HAP
emissions
generally
decrease
with
increasing
engine
load
and
operating
temperature
and
catalytic
removal
efficiency
improves
with
higher
exhaust
temperatures,
the
initial
performance
test
should
be
conducted
at
the
lower
end
of
this
range.
Completing
performance
testing
under
the
above
specified
conditions
should
result
in
the
maximum
expected
generation
of
HAP
from
the
engine.
Consequently,
the
operator
and
EPA
can
be
assured
that
HAP
emissions
will
be
lower
across
the
remaining
recommended
operating
range.
In
order
to
assure
that
RICE
will
be
able
to
continue
to
operate,
special
provisions
regarding
start­
up,
idle,
and
shutdown
operation
should
be
specified
in
the
final
rule,
and
the
commenter
stands
ready
to
work
with
EPA
to
develop
workable
solutions
and
language.

Response:
The
EPA
has
removed
the
requirement
to
maintain
an
operating
load
or
fuel
flow
rate
equal
to
or
greater
than
95
percent
of
the
operating
load
or
fuel
flow
rate
established
during
the
initial
performance
test.
This
was
discussed
in
response
to
comment
V.
B.
1.2.
Furthermore,
EPA
has
based
the
final
emission
limitations
on
test
results
from
testing
conducted
at
high
loads
only.
Since
the
emission
limitations
are
based
on
emissions
data
from
high
load
tests,
EPA
has
found
it
appropriate
to
require
that
performance
test
be
conducted
at
high
load
conditions.
This
has
been
specified
in
the
final
rule.
The
EPA
has
defined
high
load
conditions
as
100
percent
±
10
percent.
Sources
are
not
required
to
meet
the
emission
limitations
at
low
load.

VI.
B.
2.5
Comment:
One
commenter
(
549)
expressed
the
opinion
that
the
initial
performance
test
period
for
2SLB,
4SLB,
and
CI
RICE
should
be
increased
from
4
to
24
hours
to
obtain
results
representative
of
the
normal
duty
cycle.

Response:
The
EPA
does
not
agree
with
this
comment;
it
is
believed
that
4
hours
is
sufficient
time
to
conduct
the
initial
performance
test.
This
was
discussed
in
response
to
comment
V.
F.
3.

VI.
B.
2.6
Comment:
One
commenter
(
549)
asked
that
EPA
include
similar
language
as
in
89
the
Petroleum
Refinery
MACT
for
Catalytic
Cracking
Units
which
has
the
provision
to
make
adjustments
to
one
of
the
monitored
operating
parameters
to
acknowledge
that
it
may
not
be
possible
to
achieve
worst­
case
operation
during
the
performance
test.
In
this
scenario,
the
testing
of
a
similar
unit
should
be
allowed
to
serve
as
the
basis
for
establishing
acceptable
inlet
temperatures.

One
commenter
(
569)
remarked
that
initial
performance
tests
should
only
have
to
be
performed
on
one
engine
when
an
installation
is
provided
with
several
identical
engines.

Response:
The
EPA
does
not
agree
that
it
is
appropriate
to
allow
a
facility
with
identical
engines
to
conduct
testing
on
only
one
of
the
units
to
establish
operating
parameters.
Although
the
units
are
identical,
operating
parameters,
as
well
as
emissions,
could
vary
significantly
from
unit
to
unit.
The
EPA
does
not
agree
that
it
is
appropriate
to
allow
a
facility
with
identical
engines
to
conduct
performance
tests
on
only
one
of
the
units
to
demonstrate
compliance
with
the
emission
limits
for
all
of
the
identical
units.
It
is
EPA's
experience
that
emissions
from
identical
units
can
vary
significantly.

VI.
B.
2.7
Comment:
One
commenter
(
569)
stated
that
manufacturer's
performance
data
should
be
allowable
in
lieu
of
an
initial
performance
test.

Response:
The
EPA
is
not
allowing
manufacturer's
performance
data
in
lieu
of
an
initial
performance
test.
Performance
data
provided
by
the
manufacturer
may
not
be
representative
of
how
the
engine
will
perform
in
the
field
and
may
overestimate
the
engine's
performance.

VI.
B.
2.8
Comment:
One
commenter
(
567)
contended
that
the
stack
testing
should
be
no
more
frequent
than
semiannual
for
CO.
The
stack
testing
for
formaldehyde
should
be
no
more
frequent
than
annual.
The
commenter
added
that
both
should
also
include
the
ability
to
go
to
even
less
frequent
testing
based
upon
good
performance.

Response:
The
EPA
agrees
with
the
commenter
and
feels
that
it
is
appropriate
to
require
semiannual
performance
tests
for
CO
for
sources
meeting
the
CO
percent
reduction
requirement.
This
has
been
specified
in
the
final
rule.
The
rationale
for
reducing
the
CO
testing
requirement
was
previously
discussed.
For
CO
stack
testing,
EPA
also
agrees
with
the
commenter
that
it
is
appropriate
to
allow
sources
that
demonstrate
compliance
for
two
consecutive
tests,
to
reduce
the
frequency
of
subsequent
performance
tests
to
annually.
However,
if
an
annual
performance
test
indicates
a
deviation
of
CO
emissions
from
the
CO
reduction
requirement,
sources
must
return
to
semiannual
performance
tests.
Regarding
formaldehyde
testing,
EPA
disagrees
with
the
commenter
and
feels
that
it
has
appropriately
set
the
testing
requirements
for
formaldehyde
at
semiannual
performance
tests.
Periodic
stack
testing
for
CO
and
formaldehyde
will
ensure,
on
an
ongoing
basis,
that
the
source
is
meeting
the
emission
limitation
requirements.
For
formaldehyde
stack
testing,
if
you
have
demonstrated
compliance
for
two
consecutive
tests,
you
may
reduce
the
frequency
of
subsequent
performance
tests
to
annually.
However,
if
the
results
of
any
subsequent
annual
performance
test
indicate
that
the
stationary
engine
is
not
in
compliance
with
the
formaldehyde
emission
limitation,
or
you
deviate
from
any
of
your
operating
limitations,
you
must
90
resume
semiannual
performance
tests.

VI.
B.
2.9
Comment:
One
commenter
(
565)
was
of
the
opinion
that
EPA
should
allow
facilities
complying
with
the
formaldehyde
emission
limitation
to
use
existing
performance
test
data
to
demonstrate
initial
compliance
with
the
emission
limit.

Response:
The
EPA
agrees
with
the
commenter
that
existing
performance
test
data
can
be
used
to
demonstrate
compliance
with
the
emission
limit.
The
facility
must
petition
the
Administrator
for
approval
and
demonstrate
that
the
tests
were
conducted
using
the
same
test
methods
specified
in
the
subpart,
the
test
method
procedures
were
correctly
followed,
no
process
or
equipment
changes
have
been
made
since
the
test,
and
the
data
is
of
good
quality
and
is
less
than
2
years
old.
Existing
test
data
can
only
be
used
to
demonstrate
initial
compliance;
after
the
initial
compliance
demonstration,
facilities
must
then
begin
to
follow
the
annual
compliance
test
schedule.

VI.
B.
2.10
Comment:
One
commenter
(
545)
stated
that
if
EPA
does
not
exempt
all
limited
use
units
from
the
rule,
requirements
for
performance
testing
in
Table
6
of
the
proposed
rule
should
be
amended
to
provide
alternative
performance
testing
requirements
for
infrequently
operated
or
seasonally
operated
stationary
RICE.

Response:
The
frequency
of
performance
testing
might
be
as
low
as
once
per
year,
if
the
source
has
demonstrated
compliance
for
two
successive
tests.
For
4SRB
units
complying
with
the
requirement
to
reduce
formaldehyde
emissions
and
that
are
less
than
5,000
HP,
only
an
initial
performance
test
is
required.
Owners
and
operators
of
seasonally
or
infrequently
operated
stationary
RICE
can
schedule
the
performance
test
for
when
the
unit
will
be
operating.
For
these
reasons,
EPA
does
not
feel
it
is
necessary
to
establish
alternative
performance
testing
requirements
for
infrequently
operated
or
seasonally
operated
stationary
RICE.
Infrequently
operated
units
subject
to
emission
limitations
must
follow
the
same
performance
testing
requirements
as
other
units.

VI.
B.
2.11
Comment:
One
commenter
(
549)
stated
that
EPA
should
define
the
number
of
test
runs
and
duration
of
test
runs
for
periodic
CO
monitoring
using
portable
CO
analyzer.
The
RICE
MACT
should
stipulate
the
test
duration
for
periodic
testing
of
engines
using
the
referenced
portable
monitoring
protocol.
While
the
protocol
includes
quality
assurance/
quality
control
and
test
procedures,
the
duration
of
the
test
is
not
defined.
Portable
monitoring
has
been
implemented
in
a
number
of
States
to
address
monitoring
requirements
for
the
title
V
operating
permit
program.
From
a
limited
review
of
State
approaches,
it
is
apparent
that
some
States
rely
on
Conditional
Test
Methods
from
the
EPA
Emission
Measurement
Center
website,
including
CTM­
030,
which
is
the
method
from
which
the
ASTM
method
was
derived.
In
implementing
these
procedures,
the
duration
for
emissions
data
acquisition
is
as
short
as
2
minutes
(
per
CTM­
034).
Test
runs
on
the
order
of
15­
20
minutes
are
more
common.
For
example,
a
Wyoming
protocol
based
on
CTM­
030
defines
the
test
duration
as
a
minimum
of
21
minutes,
with
the
start
of
the
emissions
test
period
dependent
upon
analyzer
response
as
defined
in
the
protocol
(
and
consistent
with
CTM­
030).
We
recommend
that
EPA
follow
a
similar
approach
and
define
the
sampling
91
period
for
acquisition
of
emissions
data
as
a
minimum
of
15
minutes
for
portable
analyzer
testing.
A
single
run
would
constitute
a
test.

Response:
The
EPA
agrees
with
the
commenter
that
for
RICE
a
minimum
15
minute
run
time
should
be
specified.
However,
the
tester
must
additionally
document
conformance
with
the
ASTM
D6522­
00
performance
requirements
of
minimum
sample
points,
and
equal
sample
time
per
point,
detailed
in
section
10
of
the
method
which
may
require
longer
run
times.
If
the
three
point
option
is
used,
then
recording
five
1­
minute
values
(
after
purging
the
greater
of
two
times
the
response
time
or
the
stability
time)
at
each
point
could
minimally
constitute
a
run.
If
the
source
can
qualify
for
the
single
point
sampling
then
recording
15
1­
minute
values
(
after
purging
the
greater
of
two
times
the
response
time
or
the
stability
time)
would
be
accepted.
The
EPA
disagrees
with
the
commenter
that
a
single
run
should
constitute
a
test.
The
EPA
wants
to
require
the
same
number
of
runs
in
a
test
regardless
of
method
followed.
(
Section
63.7(
e)(
3)
of
the
General
Provisions
requires
three
unless
the
final
rule
specifies
differently.)

VII.
COST/
IMPACT
VII.
1
Comment:
One
commenter
(
555)
stated
that
NOx
increases
due
to
oxidation
catalysts
for
2SLB
and
4SLB
engines
should
be
considered
in
evaluating
the
cost
and
benefits
of
the
proposed
rule.
Test
results
for
2SLB
and
4SLB
engines
(
Docket
No.
OAR­
2002­
0059­
0049
(
A­
95­
35
II­
B­
27))
equipped
with
oxidation
catalysts
indicate
an
increase
of
NOx
emissions
up
to
about
15
percent
and
12
percent
for
2SLB
and
4SLB
engines,
respectively.
It
is
not
clear
that
the
impacts
of
this
NOx
increase
has
been
addressed
with
respect
to
the
ability
of
sources
to
comply
with
State
and
local
NOx
limits
or
impacts
on
the
environment.

Response:
The
EPA
did
consider
NOx
increases
due
to
oxidation
catalysts
for
2SLB
and
4SLB
engines.
However,
the
NOx
increases
resulting
from
2SLB
and
4SLB
installing
oxidation
catalyst
controls
to
comply
with
the
final
rule
are
far
less
than
the
NOx
decreases
resulting
from
4SRB
engines
installing
NSCR
controls
to
comply
with
the
final
rule,
resulting
in
a
net
decrease
in
NOx
emissions
due
to
the
final
rule
and
a
benefit
to
the
environment
overall.
In
addition,
oxidation
catalysts
are
not
specifically
required
by
the
final
rule
and
as
only
new
2SLB
and
new
4SLB
engines
are
affected
by
the
final
rule,
sources
that
are
concerned
about
NOx
emissions
can
use
other
methods
of
HAP
emission
control
that
are
less
problematic
from
a
NOx
control
perspective
(
like
in­
cylinder
controls),
or
they
can
use
NOx
control
to
reduce
NOx
from
engines
using
oxidation
catalysts.

VII.
2
Comment:
One
commenter
(
569)
expressed
that
the
RICE
MACT
is
not
cost
effective
for
existing
4SRB
engines.

Response:
In
the
development
of
MACT
floors,
EPA
does
not
consider
the
cost
effectiveness
of
implementing
rule
requirements.
In
developing
options
more
stringent
than
the
floor,
the
EPA
can
consider
costs
of
achieving
emissions
reductions
and
base
a
decision
on
whether
or
not
going
beyond­
the­
floor
is
cost
effective.
92
VII.
3
Comment:
One
commenter
(
570)
stated
that
it
may
be
possible
for
EPA
to
obtain
the
same
level
of
HAP
emissions
reductions
from
stationary
RICE
in
objectively
and
discretely
defined
high­
risk
urban
areas
at
significantly
lower
costs
than
will
be
realized
in
the
proposed
rule.
The
commenter
offered
a
model
which
includes
an
operation
and
maintenance
plan
(
O&
M
Plan)
for
the
control
device,
initial
performance
test,
compliance
with
O&
M
represents
compliance
with
emission
standards,
control
device
summaries
and
excess
emission
reports
to
be
provided
to
the
Administrator
on
an
annual
basis.

Response:
It
appears
the
commenter
is
suggesting
that
the
rule
cover
high­
risk
urban
areas
only.
The
EPA
is
unable
to
do
so
as
EPA
is
required
by
the
CAA
to
establish
NESHAP
and
set
MACT
requirements
for
the
source
category
and
subcategories.
In
addition,
the
commenter's
model
suggests
a
case­
by­
case
operating
plan
which
would
greatly
increase
the
regulatory
resources
needed
for
the
rule
and
could
lead
to
significantly
different
requirements
for
similar
sources.
In
addition,
as
discussed
in
response
to
several
other
comments,
EPA
feels
that
there
are
very
good
reasons
to
require
periodic
testing
of
all
regulated
sources.

VII.
4
Comment:
One
commenter
(
583)
asserted
that
if
EPA
adopts
the
new
MACT
standards
as
proposed,
the
rule
would
adversely
affect
its
members
by
increasing
the
cost
to
operate
compressor
engines
on
pipelines
and
at
natural
gas
storage
facilities,
and
by
discouraging
customers
from
using
natural
gas­
fired
engines
as
distributed
energy.

Response:
The
EPA
does
not
agree
with
the
commenter
that
the
rule
would
adversely
affect
the
regulated
community.
The
economic
impact
analysis
for
the
final
rule,
which
is
available
in
the
public
docket,
shows
that
costs
to
industries
and
changes
in
prices
of
natural
gas
will
be
less
than
1
percent
overall.
These
impact
estimates
do
account
for
the
effect
on
consumers
from
the
increase
in
natural
gas
prices,
and
account
for
the
pass
through
of
higher
natural
gas
prices
to
consumers
as
well
as
the
higher
cost
of
operations
to
natural
gas
compressor
engine
operators
and
storage
facilities.
It
should
be
also
be
noted
that
EPA
is
required
by
the
CAA
to
regulate
this
source
category.

VIII.
RISK
Note:
All
of
the
comments
pertaining
to
risk
are
summarized
below.
There
is
a
single
response
given
at
the
end
of
this
section
for
all
of
the
risk
comments.

VIII.
1
Comment:
Three
commenters
(
558,
564,
585)
supported
the
use
of
risk­
based
applicability
criteria
to
remove
sources
that
do
not
pose
significant
risk.
VIII.
2
Comment:
One
commenter
(
565)
stated
that
allowing
the
use
of
the
CAA
section
112(
d)(
4)
provision
within
the
source
category
could
provide
substantial
cost­
effectiveness
benefits
while,
at
the
same
time,
providing
protection
to
human
health
and
the
environment.
However,
the
commenters
believe
that
EPA
must
clarify
in
the
final
rule
exactly
how
the
provisions
will
be
implemented.
VIII.
3
Comment:
One
commenter
(
562)
believes
that
there
are
ways
to
structure
the
rule
93
to
focus
on
facilities
that
pose
significant
risks
and
avoid
imposition
of
high
costs
on
facilities
that
pose
little
risk.
An
appropriate
approach
would
be
to
allow
individual
facilities
to
conduct
a
risk
assessment
to
show
that
it
poses
insignificant
risks
to
the
public.
Commenter
562
stated
that
such
an
approach
would
involve
air
quality
modeling
and
the
use
of
RfCs.
VIII.
4
Comment:
One
commenter
(
570)
believes
that
the
proposed
RICE
MACT
should
be
either
withdrawn
and
stationary
RICE
de­
listed
pursuant
to
112(
c)
or
the
rule
should
be
limited
in
scope
to
those
sources
where
the
social
benefits
resulting
from
the
relatively
small
HAP
emissions
reductions
can
be
demonstrated
(
e.
g.,
large
urban
areas).
In
this
latter
case,
the
rule
should
be
greatly
simplified.
The
commenter
supported
inclusion
of
the
risk­
based
approaches
presented
in
the
preamble
but
does
not
see
how
the
RICE
category
fits
into
the
risk­
based
approach
that
EPA
wants
to
use.
VIII.
5
Comment:
Three
commenters
(
548,
560,
595)
encouraged
EPA
to
continue
considering
the
establishment
of
a
subcategory
of
RICE
that
would
meet
112(
c)(
9)(
B)'
s
riskbased
criteria
for
delisting,
and
support
the
use
of
112(
c)(
9)(
B)
to
achieve
the
goals
of
the
rule
in
a
less
costly
manner.
VIII.
6
Comment:
One
commenter
(
550)
stated
that
they
support
focusing
efforts
on
the
greatest
risks
to
public
health,
but
suggested
that
the
risk­
based
approaches
be
presented
to
an
appropriate
scientific
peer
review
group
(
e.
g.,
the
CAA
Advisory
Committee
or
Science
Advisory
Board)
before
being
placed
in
the
preamble.
VIII.
7
Comment:
Three
commenters
(
499,
572,
586)
are
opposed
to
the
risk­
based
exemptions.
Commenter
499
noted
that
the
proposal
to
include
risk­
based
exemptions
is
critically
flawed
and
opposes
adoption
of
the
risk­
based
exemptions
into
MACT.
Commenter
586
is
extremely
concerned
about
the
policy
and
technical
implications
of
the
risk­
based
exemption
proposal.
Because
of
the
flaws
with
the
proposal,
the
commenter
is
opposed
to
the
adoption
of
the
risk­
based
exemptions
to
MACT.
Commenter
574
stated
the
belief
that
the
use
of
risk­
based
concepts
to
evade
MACT
applicability
is
contrary
to
the
intent
of
the
CAA
and
is
based
on
a
flawed
interpretation
of
section
112(
d)(
4)
of
the
CAA.
The
commenter
added
that
the
CAA
requires
a
technology­
based
floor
level
of
control
and
does
not
provide
exclusions
for
risk
or
secondary
impacts
from
applying
the
MACT
floor.
Commenter
574
stated
that
from
a
practical
standpoint,
the
approaches
(
to
risk­
based
exemptions)
in
the
preamble
are
not
appropriate.
VIII.
8
Comment:
One
commenter
(
529)
is
opposed
to
risk­
based
exemptions
discussed
in
the
preamble
and
recommends
that
EPA
expeditiously
propose
and
promulgate
the
rule
without
such
exemptions.
Commenter
567
is
opposed
to
the
risk­
based
exemptions
and
called
upon
EPA
to
promulgate
the
remaining
technology­
based
MACT
standards
without
the
risk­
based
exemptions.
VIII.
9
Comment:
Four
commenters
(
499,
567,
572,
586)
believe
that
the
preambles
of
individual
rule
proposals
were
an
inappropriate
forum
for
introducing
significant
changes
in
the
way
that
MACT
standards
are
established.
Precedent­
setting
change
of
the
magnitude
that
EPA
has
raised
should
be
discussed
openly
and
carefully
with
all
affected
parties
instead
of
being
buried
in
the
preambles
of
individual
standards.
Commenter
574
stated
the
concern
that
other
parties
may
miss
commenting
on
the
riskbased
exemptions
because
they
are
contained
within
six
separate
proposals.
The
commenter
added
that
to
give
the
issue
full
consideration,
the
risk
provisions
should
not
be
adopted
within
any
of
the
final
rules
but
should
be
addressed
in
one
place,
such
as
in
revisions
to
the
General
94
Provisions
of
40
CFR
63
subpart
A.
VIII.
10
Comment:
One
commenter
(
586)
stated
that
the
preamble
discussion
of
the
riskbased
approaches
is
not
sufficient
to
allow
for
complete
public
comment
and,
therefore,
it
would
not
be
appropriate
for
EPA
to
go
directly
to
a
final
rule
(
without
reproposal)
with
any
of
the
approaches
outlined
in
the
proposal.
The
commenter
recommended
that
the
risk­
based
exemption
proposal
be
dropped
because
it
is
unacceptable.
Commenter
567
stated
that
the
use
of
sub­
categorization
and
source
category
deletions
under
CAA
section
112(
c)
have
been
implemented
several
times
since
the
MACT
program
began.
The
commenters
have
been
unable
to
comment
on
the
technical
merit
of
the
risk
analysis
employed
by
the
EPA.
Until
the
residual
risk
analysis
procedures
have
been
implemented
via
the
CAA
section
112(
f)
process,
risk
analysis
should
not
been
used
in
making
MACT
determinations
pursuant
to
CAA
section
112(
d)(
4)
and,
could
never
be
used
to
establish
a
MACT
floor.
VIII.
11
Comment:
Six
commenters
(
499,
529,
567,
572,
586,
588)
stated
that
the
proposal
to
include
risk­
based
exemptions
is
contrary
to
the
1990
CAA
Amendments
(
CAAA)
which
calls
for
MACT
standards
based
on
technology
rather
than
risk
as
a
first
step.
Congress
incorporated
the
residual
risk
program
under
CAA
section
112(
f)
to
follow
the
MACT
standards
(
not
to
replace
them).
The
need
for
the
technology­
based
approach
has
been
recently
reinforced
by
the
results
of
the
National
Air
Toxics
Assessment
(
NATA),
which
indicates
that
exposure
to
air
toxics
is
very
high
throughout
the
country
in
urban
and
remote
areas.
Commenters
586
and
499
added
that
risk­
based
approaches
will
be
used
separately
to
augment
and
improve
technology­
based
standards
that
do
not
adequately
provide
protection
to
the
public.
Commenter
572
added
that
section
112(
f)
of
the
CAA
was
developed
to
address
residual
risks
remaining
after
implementation
of
technology­
based
MACT
standards
and
was
intended
to
provide
additional
protection,
not
replace
technology
controls.
VIII.
12
Comment:
Five
commenters
(
499,
529,
567,
572,
586)
stated
that
the
proposal
to
allow
risk­
based
exemptions
would
divert
back
to
the
time­
consuming
NESHAP
development
process
that
existed
prior
to
the
CAAA.
Under
this
process,
which
began
with
a
risk
assessment
step,
only
eight
NESHAP
were
promulgated
during
a
20­
year
period.
If
the
proposed
approaches
are
inserted
into
upcoming
standards,
the
commenters
fear
the
MACT
program
(
which
is
already
far
behind
schedule)
would
be
further
delayed.
VIII.
13
Comment:
Three
commenters
(
499,
567,
586)
stated
that
the
risk­
based
exemption
proposal
removes
the
"
level­
playing
field"
that
would
result
from
the
proper
implementation
of
technology­
based
MACT
standards.
Establishing
a
baseline
level
of
control
is
essential
to
prevent
industry
from
moving
to
areas
of
the
country
that
have
the
least
stringent
air
toxics
programs,
which
was
one
of
the
primary
goals
of
developing
a
uniform
national
air
toxics
program
under
section
112
of
the
1990
CAA
amendments.
The
risk­
based
approaches
would
jeopardize
future
reductions
of
HAP
in
a
uniform
and
consistent
manner
across
the
nation.
Commenter
499
stated
that
the
NATA
data
show
that
virtually
no
area
of
the
country
has
escaped
measurable
concentrations
of
toxic
air
pollution.
The
NATA
information
indicates
that
exposure
to
air
toxics
is
high
in
both
densely
populated
and
remote
rural
areas.
VIII.
14
Comment:
Four
commenters
(
548,
560,
565,
595)
believe
that
CAA
section
112(
d)(
4)
provides
EPA
with
authority
to
exclude
sources
that
emit
threshold
pollutants
from
regulation.
The
commenters
indicated
that
CAA
section
112(
d)(
4)
allows
for
discretion
in
developing
MACT
standards
for
HAP
with
health
thresholds.
This
is
consistent
with
the
plain
95
language
of
the
statute,
which
states
that:
"
With
respect
to
pollutants
for
which
a
health
threshold
has
been
established,
the
Administrator
may
consider
such
threshold
level,
with
an
ample
margin
of
safety,
when
establishing
emission
standards
under
this
subsection."
The
use
of
CAA
section
112(
d)(
4)
authority
also
is
supported
by
CAA's
legislative
history,
which
emphasizes
that
Congress
included
§
112(
d)(
4)
in
the
CAA
to
prevent
unnecessary
regulation
of
source
categories.
Multiple
commenters
(
548)
referenced
Sen
Rep.
101­
228,
at
176
(
1989),
reprinted
in
1990
U.
S.
C.
C.
A.
N.
3385,
3560:
"[
W]
here
some
sources
do
emit
more
than
the
threshold
amount,
the
Administrator
is
authorized
by
section
112(
d)(
4)
to
use
the
no
observable
effects
level
or
NOEL
(
again
with
an
ample
margin
of
safety)
as
the
emission
limitation
in
lieu
of
more
stringent
"
best
technology"
requirements.
Following
this
scenario,
only
those
sources
in
the
category
which
present
a
risk
to
public
health
(
those
emitting
in
amounts
greater
than
the
safety
threshold)
would
be
required
to
install
controls,
even
though
the
general
policy
is
"
maximum
achievable
technology
everywhere."
Again,
there
is
a
means
to
avoid
regulatory
costs
which
would
be
without
public
health
benefit."
VIII.
15
Comment:
Four
commenters
(
547,
548,
560,
595)
supported
the
use
of
§
112(
d)(
4)
applicability
cutoffs
for
both
threshold
and
non­
threshold
pollutants.
Commenter
548
stated
that
the
plain
language
of
112(
d)(
4)
does
not
distinguish
between
carcinogens
and
non­
carcinogens.
Commenter
595
stated
that
nothing
in
CAA
section
112(
d)(
4)
limits
it
to
non­
carcinogens,
and
EPA
should
consider
112(
d)(
4)
whenever
it
sets
emission
standards
for
HAP
regardless
of
health
endpoint.
Commenters
548,
595,
and
560
stated
that
advances
in
risk
assessment
science
indicate
that
some
carcinogens
may
be
threshold
pollutants.
Commenter
547
believes
that
a
health
threshold
of
one
in
one
million
cancer
risk
is
appropriate
for
non­
threshold
pollutants.
VIII.
16
Comment:
One
commenter
(
565)
believes
that
EPA
could
implement
a
CAA
section
112(
d)(
4)
emissions
limitation
under
both
the
first
and
third
scenarios
discussed
in
the
preamble
(
scenario
1:
exempt
low
risk
facilities
emitting
only
threshold
pollutants,
scenario
3:
exempt
emission
points
at
facilities
that
emit
only
threshold
pollutants).
However,
the
commenters
believe
that
the
use
of
a
CAA
section
112(
d)(
4)
emissions
limit
as
described
under
the
third
scenario
in
the
preamble
would
provide
the
maximum
benefit
of
the
CAA
section
112(
d)(
4)
provision.
Under
this
scenario,
facilities
that
emit
both
threshold
and
nonthreshold
pollutants
could
achieve
exemption
from
MACT
controls
for
threshold
HAP
emission
points
based
on
their
ability
to
meet
the
associated
health
threshold
for
those
HAP.
Another
possible
use
of
the
CAA
section
112(
d)(
4)
emissions
limitation
that
EPA
discussed
would
apply
to
both
threshold
and
nonthreshold
pollutants.
VIII.
17
Comment:
Two
commenters
(
574,
575)
disagreed
that
§
112(
d)(
4)
can
be
interpreted
to
allow
exemptions
for
individual
sources.
The
commenters
believe
that
§
112(
d)(
4)
applies
to
categories
of
sources.
Commenter
575
understood
CAA
section
112(
d)(
4)
to
allow
the
Agency
to
reduce
the
stringency
of
a
MACT
standard
applicable
to
an
entire
category
or
subcategory
based
on
a
health
threshold
for
a
particular
pollutant.
VIII.
18
Comment:
One
commenter
(
567)
stated
that
the
plain
meaning
of
§
112(
d)(
4)
96
does
not
allow
EPA
to
make
MACT
standard
exemptions
for
individual
sources.
Commenter
567
believes
that
EPA
has
misinterpreted
the
provision
in
112(
d)(
4).
Section
112(
d)(
4)
of
the
CAA
does
not
state
that
EPA
can
use
applicability
thresholds
"
in
lieu
of"
the
CAA
section
112(
d)(
3)
MACT
floor
requirements.
The
commenters
interpreted
CAA
section
112(
d)(
4)
to
state
that
health
based
thresholds
can
be
considered
when
establishing
the
degree
of
the
MACT
floor
requirements,
but
it
should
not
be
used
to
supplant
the
requirements
established
pursuant
to
CAA
section
112(
d)(
3).
VIII.
19
Comment:
One
commenter
(
574)
stated
that
CAA
section
112(
d)(
4)
does
not
apply
for
source
categories
that
emit
carcinogens.
VIII.
20
Comment:
One
commenter
(
586)
stated
that
the
concept
of
cancer
exposure
below
a
threshold
is
untried,
and
Congress
clearly
intended
that
carcinogens
be
considered
nonthreshold
pollutants.
VIII.
21
Comment:
One
commenter
(
565)
believes
that
EPA
could
accomplish
the
same
outcome
as
a
CAA
section
112(
d)(
4)
emissions
limitation
if
it
applied
an
emissions
limitation
on
nonthreshold
pollutants
using
its
inherent
de
minimis
authority
instead
of
a
limit
based
on
CAA
section
112(
d)(
4).
VIII.
22
Comment:
Regarding
the
use
of
a
concentration­
based
applicability
threshold,
commenter
567
stated
that
this
is
not
what
Congress
intended
in
the
CAAA
of
1990.
Congress
mandated
that
the
MACT
floor
be
established
as
initial
level
of
control.
VIII.
23
Comment:
One
commenter
(
565)
stated
that
in
the
preambles,
EPA
expresses
uncertainty
over
whether
it
has
the
authority
to
subcategorize
source
categories
based
on
risk.
The
commenter
believes
that
EPA
has
ample
authority,
based
on
CAA
sections
112(
c)(
1)
and
112(
d)(
1),
to
subcategorize
based
on
risk.
Section
112(
c)(
1)
of
the
CAA
states:
"
Nothing
in
the
preceding
sentence
[
relating
to
following
the
NSPS
program
categories
and
subcategories]
limits
the
Administrator's
authority
to
establish
subcategories
under
this
section
as
appropriate."
Thus,
Congress
allowed
EPA
discretion
to
subcategorize
previously
created
categories,
regardless
of
the
criteria
that
EPA
used
to
create
the
category
in
the
first
place,
and
to
do
so
at
any
time.
Section
112(
d)(
1)
of
the
CAA
provides
that
EPA
"
may
distinguish
among
classes,
types
and
sizes
of
sources"
when
establishing
MACT
standards.
The
broad
terms
"
classes,"
"
types,"
and
"
sizes"
indicate
that
Congress
intended
that
EPA
have
broad
discretion
in
establishing
subcategories
and
do
not
preclude
EPA
from
subcategorizing
based
on
risk,
since
low­
risk
sources
could
be
considered
a
"
class"
or
"
type"
of
source.
The
commenter
added
that
the
only
case
to
clarify
this
statutory
language
recognized
the
broad
discretion
it
confers
on
EPA
to
create
subcategories
with
different
emission
standards.
Sierra
Club
v.
Costle,
657
F.
2D
298
(
D.
C.
Cir.
1981).
The
Court
noted,
"[
t]
he
required
finding
that
must
underlie
a
variable
standard
is
much
broader
than
a
mere
determination
that
uniformity
is
not
achievable."
Id.
at
321.
On
this
basis,
the
Court
expressly
upheld
EPA's
subcategorization
of
coal­
fired
power
plants
based
on
the
sulfur
content
of
fuel.
More
generally,
the
Sierra
Club
decision
confirms
EPA's
discretion
to
set
differentiated
emissions
standards
for
subcategories
of
sources,
even
in
instances
where
the
strictest
standard
may
be
achievable
by
all
sources.
VIII.
24
Comment:
One
commenter
(
575)
did
not
see
legal
basis
for
creating
a
subcategory
based
on
risk
alone.
VIII.
25
Comment:
One
commenter
(
574)
stated
that
the
only
option
that
appears
consistent
with
the
CAA,
does
not
create
excessive
work
for
State
and
local
agencies,
and
may
be
97
able
to
be
based
on
science,
is
the
subcategorization
and
delisting
approach.
However,
the
commenter
added
that
the
subcategories
should
be
based
on
equipment
or
fuel
use,
not
risk.
The
commenter
added
that
a
subcategory
based
on
site­
specific
risk
creates
a
circular
definition
and
does
not
make
sense.
The
commenter
also
stated
that
subcategory
de­
listing
should
occur
before
the
compliance
date
so
that
facilities
do
not
put
off
compliance
in
the
hope
or
anticipation
of
delisting
VIII.
26
Comment:
One
commenter
(
575)
urged
caution
in
the
effort
to
define
a
permissible
subcategory
based
upon
technological
differences
that
would
then
be
susceptible
to
delisting
under
the
extremely
stringent
delisting
requirements
of
CAA
section
112.
VIII.
27
Comment:
One
commenter
(
574)
stated
that
the
preamble
discussion
of
a
lowrisk
subcategory
on
the
MACT
floors
for
the
entire
category
sounds
like
another
valid
reason
not
to
mix
the
risk­
based
and
technology­
based
standards
development.
The
commenter
added
that
EPA
does
not
address
how
the
"
once
in,
always
in"
policy
would
apply.
VIII.
28
Comment:
One
commenter
(
565)
supported
the
concept
described
in
the
preambles
regarding
the
establishment
of
the
MACT
floor
based
on
the
controls
for
the
entire
source
category.
The
commenter
agreed
that
considering
controls
for
low­
risk
subcategories
could
maintain
the
appropriate
stringency
of
the
MACT
floor.
Once
the
floor
is
established,
facilities
could
demonstrate
their
inclusion
in
the
low­
risk
category
that
is
subsequently
listed.
VIII.
29
Comment:
One
commenter
(
605)
agreed
that
the
MACT
floor
could
be
established
for
the
entire
source
category
and
then
facilities
could
be
allowed
to
become
part
of
a
low­
risk
subcategory
in
the
future,
after
MACT
standards
are
established.
This
would
allow
low
risk
facilities
to
use
CAA
section
112(
c)(
9)
without
affecting
the
MACT
floor.
VIII.
30
Comment:
One
commenter
(
579)
stated
that
EPA
should
add
the
reference
values
for
acute
exposure
to
Table
2
on
page
77848
of
the
proposed
rule
so
that
facilities
will
be
able
to
more
accurately
calculate
their
hazard
indices
when
the
health
risk
assessment
guidance
is
implemented.
VIII.
31
Comment:
One
commenter
(
567)
agreed
with
EPA's
choice
to
derive
their
data
from
IRIS,
CAL­
EPA
and
ATSDR
for
its
documentation
for
establishing
risk­
based
threshold
and
non­
threshold
values.
The
commenter
added
that
almost
all
HAP
are
being
reviewed
and
reevaluated
on
a
regular
basis,
and
it
would
be
inappropriate
to
single
out
formaldehyde
and
acetaldehyde
at
this
time.
The
EPA
can
only
rely
on
what
is
currently
published
and
underwent
either
peer
review
or
agency
review.
The
issue
of
changing
health­
based
guideline
values
will
always
be
a
concern
once
health­
based
regulations
are
promulgated.
VIII.
32
Comment:
One
commenter
(
574)
stated
that
EPA
should
consider
formaldehyde
and
acetaldehyde
as
carcinogens
unless
a
reassessment
classifies
them
as
threshold
pollutants.
VIII.
33
Comment:
One
commenter
(
565)
stated
that
making
an
allowance
for
other
exposures
under
CAA
section
112
is
not
necessary
to
protect
public
health.
The
commenter
added
that
consideration
of
exposures
from
other
sources
places
a
disproportionate
burden
on
major
sources.
Legislative
history
does
not
support
the
consideration
of
exposures
from
other
source
types
when
setting
risk­
based
criteria.
The
commenter
stated
that
multi­
pathway
risk
assessment
should
be
required
only
for
those
HAP
that
have
the
potential
for
causing
significant
multi­
pathway
exposure.
VIII.
34
Comment:
One
commenter
(
527)
stated
that
there
is
some
question
about
whether
formaldehyde
produced
by
natural
gas
engines
can
be
considered
as
a
significant
direct
98
source
of
formaldehyde
to
which
people
are
exposed.
An
EPA
report
by
D.
Luecken
estimates
that
85­
99
percent
of
the
aldehydes
in
the
atmosphere
are
due
to
atmospheric
formation,
not
emissions.
Commenter
527
stated
that
the
population
density
adjacent
to
major
sources
of
HAP
should
be
considered.
A
more
reasonable
rule
for
formaldehyde
should
also
consider
the
location
of
the
source
relative
to
current
population
and
possibly
a
reasonable
expectation
of
future
population
in
the
area.
VIII.
35
Comment:
One
commenter
(
565)
stated
that
uncertainty
is
already
considered
in
the
establishment
of
reference
concentrations
from
which
the
hazard
index
(
HI)
is
derived.
The
commenter
stated
that
the
uncertainty
factors
used
in
the
NATA
are
large,
and
because
of
the
considerable
uncertainty
adjustments
that
are
already
applied,
it
is
highly
unlikely
that
an
ample
margin
of
safety
would
ever
have
to
include
more
uncertainties
than
are
already
incorporated
in
the
RfCs.
In
some
cases,
the
uncertainty
corrections
are
too
conservative.
VIII.
36
Comment:
One
commenter
(
548)
stated
that
EPA
should
use
reference
concentrations
(
RfCs)
and
reference
doses
(
RfDs)
as
the
threshold
for
deciding
whether
a
MACT
standard
for
a
specific
pollutant
is
warranted.
The
RfCs
and
RfDs
are
set
conservatively
by
first
determining
a
NOEL
and
then
reducing
that
level
by
an
uncertainty
factor.
This
method
of
setting
RfCs
and
RfDs
assures
that
public
health
is
protected
by
"
an
ample
margin
of
safety."
Commenter
548
added
that,
as
EPA
suggests,
the
ample
margin
of
safety
for
nonthreshold
pollutants
is
a
cancer
risk
that
does
not
exceed
one
in
one
million.
VIII.
37
Comment:
One
commenter
(
565)
stated
that
EPA
should
clarify
that
sources
wishing
to
use
the
CAA
section
112(
d)(
4)
emissions
limit
must
only
demonstrate
compliance
with
risk­
based
thresholds
for
those
HAP
listed
in
Table
2
of
the
RICE
proposal
preamble.
VIII.
38
Comment:
One
commenter
(
565)
stated
that
hazard
quotients
for
chemical
mixes
should
not
be
summed
to
determine
the
HI
unless
the
primary
effects
are
on
the
same
organ
by
the
same
mechanism;
otherwise
the
risk
will
be
overestimated.
VIII.
39
Comment:
One
commenter
(
565)
stated
that
all
risk­
related
provisions
of
CAA
section
112
should
be
guided
by
the
purpose
of
the
ample
margin
of
safety
determination
­
namely,
to
account
for
uncertainty
in
the
underlying
health
value.
A
HI
equal
to
or
greater
than
one
should
account
for
uncertainty
and
provide
an
ample
margin
a
safety.
Commenter
565
stated
that
The
Report
of
the
Commission
on
Risk
Assessment
and
Risk
Management
provides
further
guidance
and
supports
the
commenters'
contention
that
a
HI
equal
to
or
greater
than
1
should
provide
an
ample
margin
of
safety
for
a
threshold
HAP.
VIII.
40
Comment:
One
commenter
(
565)
stated
that
adoption
of
the
Drinking
Water
Program's
concept
of
a
HI
of
0.2
is
not
supportable.
The
commenter
added
that,
at
a
minimum,
before
EPA
could
import
the
drinking
water
policy
into
air
programs,
the
agency
would
need
to
evaluate
the
available
scientific
data
 
for
the
HAP
of
concern
in
each
individual
rulemaking
 
to
determine
wether
the
data
justify
a
conclusion
that
80
percent
of
the
exposures
to
those
pollutants
come
from
sources
outside
the
source
category.
VIII.
41
Comment:
One
commenter
(
574)
stated
that
the
HI
is
useful
in
evaluating
sitespecific
impacts,
but
choosing
a
generic
HI
(
some
multiple
of
1)
for
application
to
a
wide
range
of
sites
is
inappropriate.
The
commenter
added
that
selection
of
an
arbitrary
multiple
of
1
is
not
science,
does
not
conform
with
CAA
section
112(
d)(
4)
and
does
not
protect
public
health.
The
commenter
added
that
using
background
concentrations
from
NATA
and
a
HI
of
1
is
99
inappropriate
because
NATA
information
includes
warnings
that
the
information
is
useful
for
large­
scale
planning
purposes
and
not
for
local
area
assessment.
VIII.
42
Comment:
One
commenter
(
567)
evaluated
the
four
potential
options
that
EPA
proposed
to
ensure
that
a
risk
analysis
under
CAA
section
112(
d)(
4)
considers
the
total
ambient
air
concentrations
of
all
the
HAPs
to
which
the
public
is
exposed.
Option
1,
which
requires
that
the
HI
for
all
pollutants
be
no
greater
than
1,
does
not
consider
additional
sources
or
background
and
is
unacceptable.
Option
3,
which
uses
existing
data
such
as
the
National
Scale
Air
Toxics
Assessment
(
NATA)
to
determine
background
and
requires
that
the
HI
be
no
greater
than
1,
is
also
unacceptable.
The
EPA
has
clearly
stated
at
public
meetings
that
the
NATA
is
not
to
be
used
to
make
regulatory
decisions.
The
NATA
relies
on
data
submitted
to
EPA
voluntarily
and
has
been
reported
to
consistently
underestimate
measured
concentrations.
Until
EPA
requires
that
HAP
inventories
be
submitted
as
proposed
in
the
CERR,
and
the
NATA
conducts
refined
modeling
around
stationary
sources,
the
NATA
should
not
be
considered
for
estimating
background
concentrations.
Option
4,
which
allows
individual
facilities
to
monitoring
the
HAP
backgrounds
for
use
in
their
own
analysis,
will
require
oversight
and
evaluation
by
the
States
to
ensure
proper
site
selections
and
analytical
methods
and
should
not
be
considered.
The
commenters
believe
Option
2,
which
requires
that
the
HI
be
no
greater
than
0.2,
would
be
the
only
viable
option
at
this
time
using
a
conservative
risk
screening
analysis.
However,
the
commenters
did
not
endorse
using
any
of
the
proposed
threshold
limit
applicability
methods
to
exempt
process
sources
from
NESHAP
requirements.
VIII.
43
Comment:
One
commenter
(
586)
stated
that
in
the
case
of
RICE
and
boilers/
process
heaters,
which
emit
acrolien,
there
appears
to
be
no
reasonable
limit
(
that
would
allow
sufficient
protection
between
the
contribution
of
a
potentially
affected
source
and
the
existing
background
concentrations
of
non­
carcinogens)
since
exposure
to
acrolien
is
predicted
to
be
high
almost
everywhere
in
the
country
(
based
on
NATA).
Uniformly
applied
NESHAP
are
necessary
to
move
this
country
toward
lower
exposures
to
irritants
like
acrolien.
VIII.
44
Comment:
One
commenter
(
574)
stated
that
the
selection
of
a
0.2
HI
as
a
rough
screening
tool
seems
reasonable,
although
it
is
unsupported
by
any
analysis.
The
commenter
added
that
if
a
default
HI
is
used,
EPA
should
include
a
provision
that
would
disallow
the
use
to
exclude
a
facility
from
MACT,
now
or
in
the
future,
if
better
background
information
is
available
that
suggests
that
the
default
does
not
protect
public
health.
The
commenter,
however,
believes
that
the
interpretation
that
includes
the
use
of
such
a
default
to
allow
exemptions
for
individual
sources
is
not
supported
by
the
CAA,
and
the
expansion
of
the
interpretation
to
include
nonthreshold
pollutants
is
in
direct
conflict
with
section
112(
d)(
4)
of
the
CAA.
VIII.
45
Comment:
One
commenter
(
567)
stated
that
the
use
of
NATA
to
determine
background
concentrations
is
unacceptable.
The
EPA
has
clearly
stated
at
a
number
of
public
meetings
that
NATA
is
not
to
be
used
to
make
regulatory
decisions.
VIII.
46
Comment:
One
commenter
(
586)
stated
that
the
proposal
is
critically
flawed
because
risk­
based
exemptions
ignore
the
cumulative
risk
that
comes
from
exposure
to
multiple
air
toxics
sources
(
e.
g.,
hundreds
of
combustion
engines
and
boilers
within
a
city)
and
do
not
protect
public
health.
Although
many
individual
sources
may
pose
a
risk
below
a
designated
threshold,
the
accumulation
of
these
pollutants
can
be
hazardous.
Addressing
this
problem
will
require
a
general
reduction
in
air
toxics
emissions
across
large
and
small
sources,
not
just
those
sources
for
which
a
high
local
risk
can
be
demonstrated.
The
commenter
stated
that
NATA
100
indicates
that
air
toxics
exposures
are
already
high
throughout
the
country.
Commenter
586
stated
that,
because
background
risk
is
already
too
high,
the
inhalation
margin
allowed
for
an
individual
source
would
be
zero,
which
leads
back
to
Congress'
decision
that
all
HAP
sources
should
take
reasonable
(
i.
e.,
MACT)
steps
to
reduce
emissions.
VIII.
47
Comment:
One
commenter
(
574)
stated
that
when
persistent
biological
toxicant
or
metal
emissions
are
significant,
ingestion
and
other
pathways
should
be
considered
in
the
risk
screening.
VIII.
48
Comment:
One
commenter
(
567)
stated
that
analyses
like
the
concentrationbased
applicability
threshold
approach
do
not
address
non­
inhalation
exposures
or
adverse
effects
on
the
environment.
Commenter
567
stated
that
allowing
individual
facilities
to
monitor
the
HAP
backgrounds
for
use
in
their
own
analysis
would
require
oversight
and
evaluation
by
State
and
local
agencies
to
insure
proper
site
selections
and
analytical
methods
and
would
be
costly
to
administer
and,
therefore,
not
acceptable.
VIII.
49
Comment:
One
commenter
(
567)
stated,
in
response
to
EPA's
request
for
comment
on
subcategorizing
RICE
having
a
capacity
utilization
of
less
than
10
percent,
that
this
class
of
RICE
would
operate
mostly
in
the
summer
months
when
the
public
is
more
likely
to
be
impacted
by
the
emissions.
Acetaldehyde,
acrolein,
and
formaldehyde
all
have
documented
shortterm
acute
health
effects.
The
EPA
has
failed
to
identify
short­
term
health
effects
throughout
any
of
the
risk
analysis
proposals.
Any
subcategorization
of
these
engines
without
controls
is
not
protective
of
public
health.
The
commenter
added
that
any
subcategorization
of
these
RICE
should
include
a
permit
requirement
that
these
units
operate
less
than
876
hours
per
year,
and
that
this
class
of
RICE
could
have
less
onerous
monitoring
requirements
as
a
way
to
lower
costs.
NOTE:
This
comment
is
also
included
under
applicability­
emergency/
limited
use.
VIII.
50
Comment:
One
commenter
(
529)
stated
that
the
proposal
does
not
address
ecological
risk
that
may
result
from
uncontrolled
HAP
emissions,
especially
in
those
areas
with
sensitive
habitats
but
few
people
nearby
to
be
exposed.
VIII.
51
Comment:
One
commenter
(
565)
support
EPA's
proposed
tiered
modeling
approach,
which
begins
with
simple
"
look­
up
tables"
and
progresses
to
more
refined
facilityspecific
risk
assessments.
Commenter
565
stated
that
an
initial
simplified
tier
of
risk
assessment,
such
as
look­
up
table,
nomograph,
or
equivalent,
should
be
embedded
in
individual
rulemakings.
The
commenters
added
that
a
guidance
document
should
address
two
additional
tiers
of
assessment:
a
conservative
screening
approach,
and
a
flexible
refined
approach.
Commenter
565
stated
that
a
risk
assessment
guidance
document
should
not
attempt
to
address
policy
and
regulatory
decisions.
Rather,
regulatory
goals
and
policies
should
be
put
forth
within
individual
notice
and
rulemakings.
Appropriate
risk
assessment
endpoints
of
concern
will
also
be
established
within
these
rulemakings.
Commenter
565
stated
that
refined
risk
assessments
(
3rd
tier)
should
allow
for
more
accurate
estimates
of
maximum
individual
risk,
and
could
accomplish
this
through:
(
1)
modeling
ambient
exposures
to
an
actual
human
receptor
location;
(
2)
use
of
exposure
factors
or
models;
(
3)
use
of
realistic
exposure
assumptions
based
on
site­
specific
data
(
residential
tenure,
etc.);
and,
(
4)
use
of
probabilistic
analysis
of
uncertainty
and
variability.
Commenter
565
stated
that
rulemakings
that
use
source­
conducted
risk
assessment
should
101
appropriately
focus
the
initial
scope
of
the
assessment
on
the
HAP,
sources,
and
other
parameters
of
concern
through
applicability
criteria
specified
with
the
individual
rulemakings.
VIII.
52
Comment:
One
commenter
(
579)
agreed
that
a
tiered
approach,
including
lookup
tables,
is
a
good
idea
for
the
RICE
source
category.
The
commenter
also
suggested
that,
for
2nd
tier
modeling,
facilities
be
allowed
to
use
any
industry­
approved
models
for
evaluating
emissions
and
the
corresponding
health
risks,
because
EPA
approved
modeling
procedures
usually
are
too
conservative
and
tend
to
over­
predict
emissions.
VIII.
53
Comment:
One
commenter
(
526)
believes
that
EPA
should
establish
risk­
based
applicability
cutoffs
for
RICE
by
including
lookup
tables.
However,
because
of
the
conservative
assumptions
outlined
in
the
preamble,
the
commenter
believes
that
the
lookup
tables
would
be
of
very
limited
value.
In
addition,
very
few
facilities
could
probably
take
advantage
of
cutoffs
based
on
health
risks
since
standards
apply
only
to
RICE
at
major
sources.
VIII.
54
Comment:
One
commenter
(
574)
stated
that
the
State
of
Wisconsin
uses
a
tiered
approach
that
first
allows
sources
to
demonstrate
compliance
if
their
potential
emissions,
stack
height,
and
exhaust
direction
are
within
the
ranges
provided
in
conservative
lookup
tables.
The
second
tier
allows
facilities
to
provide
site­
specific
modeling
to
demonstrate
compliance
with
ambient
air
standards
at
the
property
line.
In
general,
the
tiered
approach
has
worked
well
in
Wisconsin.
VIII.
55
Comment:
One
commenter
(
567)
stated
that
if
EPA
decides
to
pursue
an
up
front
risk
analysis
approach,
it
should
not
be
a
tiered
approach.
The
development
of
generic
risk
screening
approach
under
the
CAA
section
112(
d)(
4)
framework
will
need
to
be
conservative,
and
the
use
of
a
(
non­
tiered)
conservative
approach
would
represent
the
least
cost
to
the
regulated
community
and
would
be
the
least
time
consuming
for
States
reviewing
the
facility's
application.
VIII.
56
Comment:
Two
commenters
(
564,
565)
stated
that
any
risk­
based
approaches
should
be
consistent
with
the
policies
in
development
for
the
residual
risk
rules.
Commenter
585
added
that
because
the
approaches
used
in
the
MACT
rulemaking
may
set
a
precedent
(
that
could
apply
to
the
residual
risk
program),
it
is
important
for
EPA
to
identify
and
address
all
of
the
issues
associated
with
the
risk­
based
approaches.
Commenter
565
stated
that
EPA
should
specify
many
of
the
procedures
for
facility­
specific
risk
assessments
in
guidelines
rather
than
in
the
rules,
because
procedures
set
in
rules
could
be
seen
as
precedent
setting
and
could
impact
the
residual
risk
program.
For
example,
the
commenters
believe
that
EPA
should
specify
the
threshold
levels
by
which
a
source
would
be
excluded
from
a
rule,
as
well
as
any
tier
1
lookup
tables,
in
each
individual
rule.
In
contrast,
the
specifics
for
how
sources
should
carry
out
tier
2
and
tier
3
risk
assessments
should
be
left
flexible
in
guidelines
now
under
development
by
EPA
for
the
residual
risk
program.
VIII.
57
Comment:
Two
commenters
(
586,
499)
stated
that
the
tools
needed
to
identify
sources
eligible
for
the
risk­
based
exemption
would
be
the
same
tools
necessary
for
a
CAA
section
112(
f)
residual
risk
assessment.
It
is
the
commenters'
understanding
that
these
tools
are
not
yet
ready
for
general
use.
Commenter
567
has
serious
reservations
with
EPA's
apparent
attempt
to
conduct
an
adhoc
risk
analysis
for
specific
source
categories
by
seeking
comments
on
the
specific
elements
to
be
included
in
the
risk
analysis
and
do
not
believe
these
rulemakings
are
an
adequate
forum
to
develop
this
risk
analysis
process.
The
commenter
believes
that
any
risk
analysis
conducted
by
102
the
EPA
must
adhere
to
the
risk
assessment
principles
outlined
in
the
Residual
Risk
Report
to
Congress.
VIII.
58
Comment:
Two
commenters
(
499,
550)
stated
that
the
proposal
will
place
a
very
intensive
resource
demand
on
State
and
local
agencies
to
review
source's
risk
assessments.
State
and
local
agencies
may
not
have
expertise
in
risk
assessment
methodology
or
the
resources
needed
to
verify
information
(
e.
g.,
emissions
data
and
stack
parameters)
submitted
with
each
risk
assessment.
VIII.
59
Comment:
One
commenter
(
574)
stated
that
they
are
concerned
about
the
potential
cost
and
workload
that
risk
provisions
would
place
on
permitting
authorities.
The
commenter
added
that
the
permitting
authorities
would
need
to
either
perform
or
verify
the
risk
analyses,
and
that
diverting
State
and
local
resources
to
focus
on
presumably
insignificant
sources
would
detract
from
efforts
associated
with
significant
sources.
The
commenter
pointed
out
some
of
the
specific
items
that
would
add
burden
to
the
State
and
local
agencies,
including
data
verification
for
background
concentrations
and
ongoing
assurance
that
low­
risk
facilities
remain
low­
risk.
VIII.
60
Comment:
Because
EPA
understands
the
difficulty
with
risk
assessments,
commenter
572
found
it
perplexing
that
EPA
believes
such
analyses
at
the
State
and
local
levels
would
be
an
efficient
way
to
protect
public
health.
VIII.
61
Comment:
One
commenter
(
567)
noted
that
for
EPA
to
conduct
an
up­
front
risk
analysis,
the
procedure
would
need
to
be
conducted
using
the
most
conservative
stack
parameters,
with
a
hypothetical
facility
fence
line
to
satisfy
the
many
impact
scenarios
that
could
occur.
If
EPA
intends
to
have
the
affected
industries
conduct
the
analysis,
then
EPA
must
consider
the
cost
incurred
by
States
which
may
lack
the
necessary
expertise
to
evaluate
and
review
these
analyses.
The
current
proposal
is
silent
on
these
implementation
and
cost
issues.
VIII.
62
Comment:
Two
commenters
(
586,
529)
stated
that
the
proposal
is
critically
flawed
because
implementing
the
exemption
program
will
require
significant
State
resources
to
review
risk
assessments
prepared
by
sources
trying
to
exempt
themselves
from
MACT.
This
review
will
require
expertise
in
risk
assessment
methodology
lacking
in
many
States
positioned
to
implement
MACT
standards
through
the
title
V
permit
program.
It
will
also
be
necessary
for
States
to
verify
extensive
emissions
and
stack
information
used
in
the
risk
assessment
to
ensure
that
the
assessment
has
been
done
properly.
Because
the
procedures
for
preparing
these
risk
assessments
on
a
large
scale
basis
and
for
assessing
the
potential
adverse
effects
of
the
pollutants
emitted
(
e.
g.
taking
into
account
existing
background
and
looking
for
a
threshold
level
for
carcinogens)
are
untried
and
will
require
extensive
debate
and
review
to
launch,
even
more
time
and
resources
will
be
needed.
Commenter
586
expressed
concern
about
exempting
a
facility
based
on
limited
emission
data
if
EPA
established
a
subcategory
listing
low­
risk
sources.
VIII.
63
Comment:
One
commenter
(
605)
expressed
concern
that
the
risk­
based
delisting
would
be
resource
intensive.
VIII.
64
Comment:
One
commenter
(
567)
stated
that
if
EPA
intends
to
have
the
affected
industries
conduct
the
analysis,
then
EPA
must
consider
the
additional
cost
incurred
by
smaller
sources
to
do
the
analysis.
VIII.
65
Comment:
One
commenter
(
574)
stated
that
title
V
permits
seem
to
be
the
obvious
implementation
tool,
and
that
title
V
permits
could
provide
enforceable
limitations,
appropriate
recordkeeping
requirements,
and
periodic
review
upon
renewal.
The
commenter
103
added
that
since
the
rule
would
apply
only
to
major
sources,
title
V
permits
already
are
required
and
would
not
be
an
added
burden.
The
commenter
added
that
title
V
could
be
used
to
implement
applicability
cutoffs,
but
that
the
workload
involved
with
the
options
requiring
modeling,
ambient
monitoring,
or
other
means
to
establish
background
concentrations
would
be
a
hindrance
to
any
implementation
mechanism.
The
commenter
stated
that
with
respect
to
potential
risk­
based
provisions,
monitoring
is
more
useful
for
demonstrating
non­
compliance
than
compliance
because
the
regulation
would
apply
to
potential
emissions
under
any
weather
conditions,
whereas
monitoring
reflects
current
weather
and
emission
conditions.
VIII.
66
Comment:
One
commenter
(
536)
supported
EPA
efforts
to
determine
alternative
MACT
setting
methodologies
but
strongly
recommended
that
these
be
pursued
separately
from
this
rulemaking.
This
will
provide
for
timely
issuance
of
final
RICE
and
Boiler/
Process
Heater
MACT
standards
relative
to
the
settlement
deadline.
VIII.
67
Comment:
One
commenter
(
499)
stated
that
it
is
evident
that
the
proposed
approach
to
risk­
based
exemptions
would
require
extensive
debate
and
review
in
order
to
launch,
which
will
further
delay
promulgation
of
the
remaining
MACT
standards.
Commenter
499
stated
that
delays
could
be
exacerbated
by
litigation
following
legal
challenges
to
the
rules,
and
such
delays
would
trigger
the
MACT
hammer,
which
would
unnecessarily
burden
the
State
and
local
agencies
and
the
industries.
The
commenter
concluded
that,
obviously,
further
delay
is
unacceptable.
VIII.
68
Comment:
One
commenter
(
586)
stated
that
the
risk
assessment
exemption
could
significantly
delay
compliance
with
MACT
for
sources
trying
unsuccessfully
to
opt
out
using
the
exemption.
VIII.
69
Comment:
One
commenter
(
575)
supported
efforts
to
lower
the
costs
to
industry
but
expressed
concern
about
the
legal
viability
of
the
risk­
based
approaches
discussed
in
the
proposal.
The
concerns
are
practical
and
acute,
in
that
if
the
rules
are
invalidated
the
MACT
"
hammer"
could
fall,
subjecting
sources
to
case­
by­
case
MACT
determinations.
Such
a
result
would
not
serve
the
interests
of
industry
or
any
other
stakeholders.
VIII.
70
Comment:
Two
commenters
(
529,
550)
stated
that
the
proposed
approaches
will
jeopardize
expeditious
promulgation
of
remaining
MACT
standards.
The
commenters
noted
that
according
to
a
recently
proposed
EPA
rule
regarding
CAA
section
112(
j),
the
regulated
community
and
State
and
local
agencies
would
have
to
proceed
with
part
2
permit
applications,
followed
by
case­
by­
case
MACT,
if
EPA
misses
the
newly
agreed­
upon
MACT
deadlines
by
as
little
as
2
months.
This
would
be
time
consuming,
costly,
and
burdensome
for
both
regulators
and
the
regulated
community.
VIII.
71
Comment:
One
commenter
(
567)
endorsed
the
comments
submitted
by
commenter
499.

Response:
The
preamble
to
the
proposed
rule
requested
comment
on
whether
there
might
be
further
ways
to
structure
the
rule
to
focus
on
the
facilities
which
pose
significant
risks
and
avoid
the
imposition
of
high
costs
on
facilities
that
pose
little
risk
to
public
health
and
the
environment.
Specifically,
EPA
requested
comment
on
the
technical
and
legal
viability
of
three
risk­
based
approaches:
(
1)
an
applicability
cutoff
for
threshold
pollutants
under
the
authority
of
CAA
section
112(
d)(
4);
(
2)
subcategorization
and
delisting
under
the
2­
190
authority
of
CAA
2See
68
FR
1276
(
January
9,
2003)
(
Plywood
and
Composite
Wood
Products
Proposed
NESHAP)
and
Docket
ID
No.
A­
98­
44,
Item
No.
II­
D­
525
(
White
papers
submitted
to
EPA
outlining
the
risk­
based
approaches).

104
sections
112(
c)(
l)
and
112(
c)(
9);
and
(
3)
a
concentration­
based
applicability
threshold.
2
The
EPA
indicated
that
it
would
evaluate
all
comments
before
determining
whether
either
approach
would
be
included
in
the
final
rule.
As
presented
above,
numerous
commenters
submitted
detailed
comments
on
these
risk­
based
approaches.

Based
on
EPA's
consideration
of
the
comments
received
and
other
factors,
EPA
has
decided
not
to
include
the
risk­
based
approaches
in
the
final
rule.
The
risk­
based
approaches
described
in
the
proposed
rule
and
addressed
in
the
comments
EPA
received
raise
a
number
of
complex
issues.
In
addition,
EPA
must
issue
the
final
rule
expeditiously,
because
the
statutory
deadline
for
promulgation
has
passed
and
EPA
has
agreed
to
a
binding
schedule
in
a
consent
decree
entered
in
Sierra
Club
v.
Whitman,
Civil
Action
No.
1:
01CVO1537
(
D.
D.
C.).
Given
the
range
of
issues
raised
by
the
risk­
based
approaches
and
the
need
to
promulgate
a
final
rule
expeditiously,
EPA
feels
that
it
is
not
appropriate
to
include
any
risk­
based
approaches
in
the
final
rule.
Nonetheless,
EPA
expects
to
continue
to
consider
risk­
based
approaches
in
connection
with
other
proposed
NESHAP
where
EPA
has
described
and
solicited
comment
on
such
approaches.
Finally,
while
EPA
is
not
including
risk­
based
approaches
in
the
final
rule,
EPA
has
included
a
number
of
other
measures
that
it
expects
will
reduce
the
costs
and
burdens
on
the
affected
sources.

IX.
EMISSIONS
DATA
A.
CSU
Testing
IX.
A.
1
Comment:
One
commenter
(
537)
contended
that
data
from
testing
of
2SLB
and
4SLB
should
be
disallowed.
The
commenter
provided
the
following
reasons:

(
1)
The
range
of
engine
operating
conditions
in
the
testing
of
the
2SLB
engine
and
quite
probably
the
4SLB
engine
are
far
leaner
than
the
leanest
engine
in
the
pipeline
RICE
fleet.
This
is
indicated
by
the
extremely
low
NOx
emissions.

(
2)
Engines
equipped
with
pre­
combustion
chambers
operating
extremely
lean
are
not
typical
examples
of
the
2SLB
and
4SLB
fleet.

(
3)
The
range
of
exhaust
temperatures,
air/
fuel
ratios,
and
exhaust
O
2
are
not
typical
of
2SLB
and
4SLB.

(
4)
Engines
were
laboratory
research
engines.
They
were
not
equipped
with
turbochargers,
but
with
turbocharger
simulators
that
do
not
have
the
same
traits
as
105
a
turbocharger.

(
5)
Found
no
information
in
the
piping
diagrams
of
insulation
on
the
ducting
and
manifolds
leading
from
the
engine
to
the
catalyst.
Certainly
all
ducting
is
insulated
in
industry.
The
EPA
needs
to
determine
if
any
insulation
was
in
place.

(
6)
The
following
excerpt
from
page
77840
of
the
proposed
rule
is
not
true:
"
In
general,
higher
exhaust
temperatures
lead
to
better
catalyst
performance.
This
difference
in
temperatures
is
a
function
of
the
inherent
design
of
these
engine
types
and
cannot
be
controlled
by
the
operator."
By
controlling
the
air/
fuel
ratio
of
the
engine,
the
exhaust
gas
temperature,
and
thus
the
catalyst
inlet
temperature,
can
be
precisely
controlled.

(
7)
If
HAP
data
from
the
2SLB
and
4SLB
testing
is
allowed
to
stand,
then
this
testing
must
become
the
definitive
work
on
all
pollutants
tested
as
well,
including
NOx.
The
NOx
data
should
be
forwarded
to
the
criteria
pollutant
group.

One
commenter
(
530)
disagreed
that
the
engine
at
CSU
is
representative
of
2SLB
engines
in
the
industry
due
to
low
NOx
levels,
high
levels
of
oxygen,
and
low
exhaust
temperatures.
The
2SLB
engine
was
running
considerably
leaner
than
similar
model
engines
at
similar
conditions.

Response:
The
EPA
compared
these
parameters
to
other
2SLB
and
4SLB
engines
for
which
EPA
has
information
in
the
Emissions
Database.
The
NOx
and
oxygen
levels
and
exhaust
temperatures
for
the
2SLB
and
4SLB
engines
tested
at
CSU
are
similar
to
those
observed
for
other
non­
CSU
2SLB
and
4SLB
engines
in
the
Emissions
Database.
This
analysis
is
presented
in
the
memorandum
entitled
"
Engine
Operation
at
Colorado
State
University"
available
from
EPA's
Edocket
Website
(
Docket
Number
OAR­
2002­
0059
(
A­
95­
35)).
The
EPA
feels
that
the
2SLB
and
4SLB
engines
tested
at
CSU
are
representative
of
2SLB
and
4SLB
engines
in
the
industry.
As
far
as
insulation
is
concerned,
the
catalyst
inlet
temperature
recorded
should
represent
catalyst
performance
at
that
temperature
regardless
of
insulation
presence
or
absence.
It
should
be
remembered
that
the
MACT
standard
for
new
sources
under
CAA
section
112(
d)
is
based
on
the
level
of
control
of
the
best
controlled
similar
source.

IX.
A.
2
Comment:
One
commenter
(
551)
was
concerned
that
the
limited
data
used
to
establish
the
standard
for
new
4SLB
engines
using
oxidation
catalyst
is
not
representative
of
typical
engine
operating
conditions
in
the
field
for
varied
engine
and
catalyst
manufacturers.

Response:
The
EPA
agrees
with
the
commenter
that
it
had
limited
data
but
disagrees
with
the
commenter
that
the
testing
was
not
representative.
As
explained
in
the
preamble
to
the
proposed
rule,
the
testing
conducted
at
CSU
to
obtain
HAP
and
CO
emissions
data
was
a
joint
EPA­
industry
effort.
Prior
to
testing,
EPA
and
industry
developed
a
list
of
engine
operating
parameters
that
were
known
to
vary
throughout
the
U.
S.
for
each
type
of
engine.
The
engines
and
control
devices
were
tested
at
typical
engine
conditions
in
which
these
operating
parameters
were
varied.
The
variations
in
the
emission
reduction
results
for
each
engine
type
are
due
to
the
106
variability
of
the
engine
and
control
system
and
include
a
representation
of
the
performance
of
the
best
controlled
source
for
new
engines.
Equipment
manufacturers,
catalyst
vendors,
owners
and
operators,
and
EPA
agreed
that
the
tests
conducted
at
CSU
were
representative
of
typical
engine
operating
conditions
in
the
field
for
varied
engine
and
catalyst
manufacturers.

IX.
A.
3
Comment:
One
commenter
(
530)
stated
that
the
testing
did
not
include
in
its
test
protocol
dynamic
spiking
that
is
required
in
Method
320
which
leaves
some
question
to
the
integrity
of
the
sample
measured
in
the
test
program.

Response:
An
alternative
quality
assurance
procedure
was
proposed
and
followed
resulting
in
data
of
sufficient
quality.
The
entire
FTIR
sampling
analysis
system
was
validated
on
a
2SLB
engine
by
a
dynamic
spiking
of
formaldehyde,
acrolein,
and
acetaldehyde.
The
data
were
assessed
following
Method
301
criteria.
Then,
on
a
daily
basis,
the
analyzer
was
checked
for
linearity
and
alignment,
a
diagnostic
or
transfer
standard
consisting
of
the
CO
was
used
to
confirm
accuracy,
a
second
diagnostic
standard
consisting
of
CO
2,
carbon
monoxide,
methane,
and
NOx
was
introduced
using
the
same
procedure.
Then
to
check
sampling
system
integrity,
a
formaldehyde
standard
was
introduced
directly
into
the
instrument
and
a
reading
obtained,
then
it
was
introduced
into
the
sampling
system
at
the
sample
probe
upstream
of
the
filter
and
another
reading
obtained.
The
sampling
system
pass/
fail
criterion
was
100
percent
±
10
percent
of
the
direct­
to­
the­
analyzer
reading.
Finally,
the
diagnostic
and
system
integrity
procedures
were
repeated
at
the
end
of
each
day
testing.
This
procedure
resulted
in
data
of
sufficient
quality.

B.
Other
IX.
B.
1
Comment:
One
commenter
(
552)
recommended
that
EPA
establish
an
optional
demonstration
program
that
would
allow
the
regulated
community
to
submit
data
demonstrating
the
actual
relationship
between
formaldehyde
emissions
and
CO
reduction
levels.
The
intent
of
this
compliance
option
is
to
allow
industry
to
demonstrate
compliance
with
the
proposed
formaldehyde
standard
established
in
the
rule
and
to
validate
a
corresponding
CO
reduction
percentage
for
a
particular
RICE
system.
The
CO
reduction
percentage
could
then
be
used
for
monitoring
and
compliance
purposes
instead
of
the
CO
reduction
percentage
currently
proposed
in
the
rule.
For
example,
an
engine
manufacturer
should
be
allowed
to
demonstrate
that
a
specific
4SLB
engine
model
and
associated
emissions
control
device
can
meet
the
14
ppmvd
formaldehyde
emissions
limit
at
a
corresponding
CO
reduction
rate
of
70
percent.
With
this
demonstration,
the
70
percent
CO
reduction
would
serve
as
a
reliable
and
justified
compliance
emissions
limit
for
all
operators
installing
this
engine
system,
and
the
operator
could
then
monitor
CO
reduction
to
assure
compliance.

Response:
The
EPA
feels
that
the
CO
percent
reduction
may
not
be
an
appropriate
measure
for
a
source
that
complies
with
the
formaldehyde
concentration
emission
limitation.
Carbon
monoxide
reduction
may
not
correspond
closely
to
formaldehyde
concentration,
depending
on
the
circumstances,
and
should
therefore
not
be
allowed
to
show
compliance
with
the
formaldehyde
concentration
emission
limitation.
The
reduction
of
CO
does
however
relate
to
the
reduction
of
HAP,
which
is
why
EPA
has
included
a
compliance
option
for
lean
burn
and
CI
107
engines
in
the
final
rule
the
option
to
comply
with
the
CO
percent
reduction
standard.
Another
reason
why
it
would
not
be
appropriate
to
allow
sources
to
conduct
their
own
demonstration
program
to
demonstrate
the
relationship
between
formaldehyde
emissions
and
CO
reduction
levels,
is
that
such
demonstration
programs
could
lead
to
different
requirements
for
different
engines.
In
addition,
in
the
proposed
rule,
EPA
specified
that
2SLB,
4SLB,
and
CI
engines
had
to
either
meet
the
CO
emission
reduction
requirement
if
using
an
oxidation
catalyst,
or
the
formaldehyde
concentration
requirement
if
using
means
other
than
an
oxidation
catalyst.
In
the
final
rule,
EPA
is
allowing
flexibility;
sources
can
meet
either
emission
limitation
regardless
of
which
control
technique
is
applied.
For
lean
burn
and
CI
engines
that
means
that
anyone
can
comply
with
the
CO
percent
reduction
requirement
or
the
formaldehyde
concentration
emission
limitation.
Sources
must
test
the
formaldehyde
concentration
and
show
that
they
meet
the
applicable
formaldehyde
concentration
limit,
if
that
is
the
emission
limitation
they
wish
to
demonstrate
compliance
with.

X.
OTHER
A.
General
Provisions
X.
A.
1
Comment:
One
commenter
(
565)
said
that
it
is
not
clear
whether
§
§
63.5(
a),
63.5(
b)(
3)
and
63.5(
b)(
4)
applies
to
the
construction
or
reconstruction
of
exempted
RICE
units
under
§
63.6590(
b).
There
is
a
conflict
in
requirement
indication
between
Table
8,
item
5
of
the
proposed
rule
and
§
§
63.6590(
b)
and
63.6596(
c).
The
commenter
supported
excluding
exempted
units
from
these
requirements
in
the
General
Provisions.

Response:
The
units
listed
under
§
63.9590(
b)
of
the
proposed
rule
are
not
exempted
from
the
rule.
The
units
listed
in
this
section
are
stationary
RICE
with
limited
requirements
in
the
rule,
but
these
units
are
not
exemptions.
Stationary
RICE
less
than
or
equal
to
500
HP
are
not
covered
by
the
final
rule.
This
was
discussed
in
response
to
comment
I.
A.
2
and
the
final
rule
has
been
written
accordingly.
The
EPA
believes
the
commenter
is
referring
to
§
63.9595(
c)
and
not
§
63.6596(
c),
which
is
a
section
that
does
not
exist.
Sections
63.5(
a),
63.5(
b)(
3),
and
63.5(
b)(
4)
apply
to
the
units
listed
in
§
63.6590
of
the
final
rule.

X.
A.
2
Comment:
One
commenter
(
565)
stated
that
derivative
General
Provisions
requirements
that
depend
on
a
SSMP
should
be
deemed
"
not
applicable."

Response:
The
rule
requires
sources
to
develop
and
maintain
a
SSMP.
Therefore,
derivative
General
Provisions
requirements
that
depend
on
a
SSMP
should
not
be
deemed
"
not
applicable."

X.
A.
3
Comment:
One
commenter
(
565)
requested
that
EPA
clarify
that
applicability
determinations
are
not
required
for
exempted
units.
There
is
a
conflict
between
the
indications
in
Table
8,
item
75
of
the
proposed
rule
and
in
particular
§
63.10(
b)(
3),
and
the
recordkeeping
exclusion
in
§
63.6590(
b).
Also,
§
63.6595(
c)
may
be
in
conflict
with
the
exclusion
in
108
§
63.6590(
b).

Response:
The
EPA
believes
the
commenter
is
referring
to
the
units
listed
in
§
63.6590(
b).
These
engines
are
not
exempted
units,
but
are
units
with
limited
requirements
in
the
rule.
The
EPA
disagrees
with
the
commenter.
Applicability
determinations
are
required
for
these
units.

X.
A.
4
Comment:
One
commenter
(
565)
asked
that
EPA
clarify
retesting
requirements
on
new
sources.
Section
63.6610
is
ambiguous
on
the
General
Provisions
requirement
for
some
new
sources
to
retest
3
years
after
promulgation
in
§
63.7(
a)(
2)(
ix).
Table
8,
item
24,
of
the
proposed
rule
does
not
clarify
the
issue.

Response:
Section
63.7(
a)(
2)(
ix)
of
the
General
Provisions
discusses
performance
test
dates
if
the
promulgated
standard
is
more
stringent
that
the
proposed
standard.
Sources
that
commenced
construction
or
reconstruction
between
the
proposal
and
promulgation
have
the
option
to
demonstrate
compliance
with
either
the
proposed
or
the
promulgated
standard.
If
the
owner
or
operator
chooses
to
comply
with
the
proposed
standard
initially,
the
owner
or
operator
must
conduct
a
second
performance
test
within
3
years
to
demonstrate
compliance
with
the
promulgated
standard.
Since
the
promulgated
standard
in
some
cases
is
more
stringent
than
the
proposed
standard,
EPA
has
specified
in
§
63.6610(
c)
of
the
final
rule
that
sources
that
commenced
construction
or
reconstruction
between
the
proposal
and
promulgation
have
this
option.

X.
A.
5
Comment:
One
commenter
(
565)
stated
that
§
63.1(
b)(
2)
is
a
reserved
paragraph.

Response:
The
commenter
is
correct
that
§
63.1(
b)(
2)
is
a
reserved
paragraph.
However,
EPA
does
not
feel
it
is
necessary
to
list
all
the
individual
sections
of
§
63.1
in
Table
8
of
the
final
rule,
and
has
therefore
not
noted
in
Table
8
of
the
final
rule
that
§
63.1(
b)(
2)
is
a
reserved
paragraph.

X.
A.
6
Comment:
One
commenter
(
565)
provided
comments
on
§
63.1(
b)(
3):
Except
§
63.1(
b)(
3)
does
not
apply
to
exempted
units
per
§
§
63.6590(
b)(
1)
and
(
2).
The
commenter
also
noted
conflict
of
§
§
63.6590(
b)(
1)
and
(
2)
with
§
63.6595(
c)
that
is
created
by
this
item
in
Table
8
of
the
proposed
rule.

Response:
The
units
listed
under
§
63.6590(
b)
are
not
exempted.
These
units
are
stationary
RICE
with
limited
requirements
in
the
rule.
Section
63.1(
b)(
3)
applies
to
the
units
listed
in
§
63.6590(
b).

X.
A.
7
Comment:
One
commenter
(
565)
had
a
comment
regarding
§
63.1(
c)(
1):
Subpart
ZZZZ
should
clarify
the
applicability
of
each
paragraph
of
subpart
A
to
sources
subject
to
subpart
ZZZZ.
Section
63.1(
c)(
2)
regarding
title
V
permit
requirements
for
area
sources
should
not
apply
to
the
subpart
since
area
sources
are
not
subject
to
subpart
ZZZZ.
Sections
63.1(
c)(
3)­(
4)
are
reserved
paragraphs.
109
Response:
The
EPA
does
not
think
it
is
necessary
to
clarify
the
applicability
of
each
paragraph
in
Table
8
of
the
final
rule.
The
EPA
feels
the
manner
in
which
the
sections
are
presented
is
sufficient.

X.
A.
8
Comment:
One
commenter
(
565)
referred
to
§
63.1(
c)(
5)
where
the
commenter
stated
that
EPA
should
clarify
§
63.1(
c)(
5)
does
not
apply
if
the
affected
source
only
includes
exempt
RICE
units
in
§
§
63.6590(
b)(
1)
and
(
2).
A
reference
to
this
exclusion
should
also
be
added
at
§
63.6595(
b).
Otherwise,
EPA
should
add
§
63.1(
c)(
5)
as
a
required
notification
in
§
§
63.6590(
b)(
1)
and
(
2).

Response:
The
stationary
RICE
listed
in
§
§
63.6590(
b)(
1)
and
(
2)
of
the
final
rule
are
not
exempt
units,
but
are
units
with
limited
requirements
in
the
rule.
The
units
listed
in
§
63.6590(
b)(
3)
of
the
final
rule
are
covered
by
the
rule
but
have
no
specific
requirements.
Section
63.1(
c)(
5)
therefore
applies
to
these
sources
as
well
as
to
other
affected
sources
with
emission
limitations
and
other
requirements.
The
EPA
has
not
made
the
change
recommended
by
the
commenter.

X.
A.
9
Comment:
One
commenter
(
565)
had
several
comments
regarding
the
General
Provisions
table
of
the
proposed
rule
and
provided
the
following
suggested
revisions:

§
63.4:
EPA
should
clarify
conversion
of
a
rich
burn
RICE
to
a
lean
burn
is
not
circumvention.

§
63.5(
a):
Clarify
"
Except
for
exempted
units
in
§
63.6590(
b)."

§
63.5(
b)(
3):
Clarify
"
Except
for
exempted
units
in
§
63.6590(
b)."

§
63.5(
b)(
4):
Clarify
"
Except
for
exempted
units
in
§
63.6590(
b)."

§
63.5(
d):
Clarify
"
HAP
identification
is
required
only
for
formaldehyde
under
subpart
ZZZZ."

§
63.5(
e):
Clarify
"
Note:
Engine
substitution
does
not
constitute
construction
or
reconstruction."

§
63.6(
c)(
5):
The
EPA
should
incorporate
the
default
in
this
paragraph
into
the
RICE
rule
(
3
year
compliance
period).

§
63.6(
e)(
1)­(
2):
Clarify,
"
Yes,
except
there
is
no
duty
to
comply
with
an
emission
limitation
during
SSM."
Note
that
(
2)
is
a
reserved
paragraph.

§
63.6(
e)(
3):
Agrees
but
notes
discrepancies
with
other
proposal
citations.

§
63.7(
e)(
2):
OK.
Note
the
label
for
this
paragraph
is
the
same
as
the
preceding
110
paragraph.

§
63.8(
a)(
4):
Subject
should
be
changed
to
monitoring
for
control
devices.

§
63.8(
c)(
1)(
iii):
This
paragraph
refers
to
SSMP
for
CEMS
and
points
back
to
§
63.6(
e)(
3)
for
which
the
answer
above
is
No.
The
EPA
should
change
answer
here
to
"
No."

§
63.9(
b)(
1)­(
5):
OK
but
(
3)
is
reserved.

§
63.9(
h)(
1)­(
6):
OK
but
(
4)
is
reserved.

§
63.10(
b)(
2)(
i)­(
v):
The
EPA
should
clarify
that
SSMPs
are
not
required
and
should
clarify
that
§
63.10(
b)(
2)(
iv)
and
(
v)
are
not
applicable.

§
63.10(
c):
OK
except
(
c)(
2)­(
4)
and
(
9)
are
reserved
and
(
15)
is
not
applicable
since
it
refers
to
an
SSM
plan.

§
63.10(
d)(
5):
A
No
response
to
§
63.10(
d)(
5)(
ii)
(
immediate
reporting)
should
be
indicated
consistent
with
EPA's
not
requiring
an
SSMP.

§
63.10(
e)(
3):
§
63.10(
3)(
i)(
C)
is
reserved.
The
commenter
requests
a
No
response
to
§
63.10(
e)(
3)
consistent
with
the
comment
to
substitute
work
practice
requirements
for
operating
limitations.

Response:
Rich
burn
is
defined
by
the
manufacturer's
intent.
A
conversion
of
a
rich
burn
engine
to
a
lean
burn
engine
would
therefore
not
change
the
classification
of
the
engine
for
purposes
of
the
final
rule,
unless
the
conversion
took
place
prior
to
December
19,
2002
and
the
engine
was
modified
with
passive
emission
control
technology
for
NOx
(
such
as
pre­
combustion
chambers).
The
EPA
has
written
the
definition
of
a
rich
burn
engine
in
the
final
rule
as
"
any
fourstroke
spark
ignited
engine
where
the
manufacturer's
recommended
operating
air/
fuel
ratio
divided
by
the
stoichiometric
air/
fuel
ratio
at
full
load
conditions
is
less
than
or
equal
to
1.1.
Engines
originally
manufactured
as
rich
burn
engines,
but
modified
prior
to
December
19,
2002
with
passive
emission
control
technology
for
NOx
(
such
as
pre­
combustion
chambers)
will
be
considered
lean
burn
engines.
Also,
existing
engines
where
there
are
no
manufacturer's
recommendations
regarding
air/
fuel
ratio
will
be
considered
a
rich
burn
engine
if
the
excess
oxygen
content
of
the
exhaust
at
full
load
conditions
is
less
than
or
equal
to
2
percent."

As
previously
indicated,
the
units
listed
in
§
63.6590(
b)
are
not
exempt
units
but
are
stationary
RICE
with
limited
requirements.
These
units
are
affected
sources
and
are
subject
to
the
rule.
Sections
63.5(
a),
63.5(
b)(
3),
and
63.5(
b)(
4)
apply
to
the
units
listed
in
§
63.6590(
b).

The
EPA
disagrees
with
the
commenter
that
it
should
clarify
that
HAP
identification
is
required
only
for
formaldehyde
under
subpart
ZZZZ
in
Table
8
of
the
final
rule.
Section
63.5(
d)(
1)(
ii)(
H)
111
of
the
General
Provisions
requires
that
the
type
and
quantity
of
HAP
emitted
by
the
sources,
be
included
in
the
application
for
approval
or
construction
or
reconstruction.
Sources
should
use
the
units
and
averaging
times
specified
for
formaldehyde
for
all
HAP
in
the
identification
of
HAP
expected
to
be
emitted.
The
requirement
of
section
63.5(
d)
is
consistent
with
requirements
implemented
for
other
promulgated
rules.

Regarding
the
comment
on
§
63.5(
e),
EPA
contacted
the
commenter
for
clarification.
The
commenter
stated
that
its
suggestion
for
this
clarifying
note
refers
to
its
position,
explained
in
comment
I.
A.
5,
that
it
feels
that
an
engine
that
was
built
before
December
19,
2002
should
be
considered
an
existing
source
even
though
it
may
be
installed
after
that
date
and
could
be
deemed
a
new
source.
The
commenter
stated
that
the
reason
for
suggesting
that
such
units
be
considered
existing
sources,
is
that
facilities
may
have
ordered
and
assembled
engines
before
December
19,
2002,
that
more
closely
resemble
the
technology
EPA
evaluated
for
setting
the
existing
MACT
floor,
and
thus
these
engines
could
become
obsolete
simply
because
they
have
been
kept
in
reserve
waiting
to
be
used.
The
commenter
used
the
word
"
engine
substitution"
as
a
way
of
avoiding
the
word
reconstruction,
which
has
a
particular
legal
connotation.
The
EPA
feels
that
this
comment
was
sufficiently
addressed
in
response
to
comment
I.
A.
5.
The
EPA
also
feels
that
any
replacement
engine
needs
to
meet
the
full
requirements
of
the
final
rule
applicable
to
that
engine,
because
even
similar
engines
can
have
different
emission
characteristics.

The
EPA
agrees
with
the
commenter
regarding
§
63.6(
c)(
5)
and
feels
it
is
appropriate
to
provide
existing
area
sources
that
become
major
sources
3
years
to
comply
with
the
final
rule.
This
has
been
specified
in
the
final
rule
in
§
63.6595(
b).
Any
new
or
reconstructed
stationary
RICE
must
be
in
compliance
with
the
final
rule
when
the
area
source
becomes
a
major
source
of
HAP.

Regarding
§
63.6(
e)(
1)
the
EPA
agrees
with
the
commenter
but
does
not
feel
it
is
necessary
to
clarify
this
in
Table
8
of
the
final
rule.
The
EPA
has
noted
in
Table
8
of
the
final
rule
that
§
63.6(
e)(
2)
is
a
reserved
paragraph.
The
EPA
acknowledges
that
§
63.6(
e)(
3)
was
inconsistent
with
other
citations
in
the
proposal.
However,
EPA
does
not
agree
with
the
commenter
that
a
SSMP
is
not
required.
The
EPA
has
written
Table
8
of
the
final
rule
to
indicate
that
a
SSMP
is
required.
The
subject
in
Table
8
of
the
final
rule
for
§
63.7(
e)(
2)
is
"
Conduct
of
performance
tests
and
reduction
of
data."
The
EPA
agrees
with
the
commenter
regarding
the
subject
of
§
63.8(
a)(
4),
which
will
be
"
Monitoring
for
control
devices."
The
EPA
disagrees
with
the
commenter
that
the
answer
to
§
63.8(
c)(
1)(
iii)
should
be
"
No"
instead
of
"
Yes."
As
previously
indicated,
EPA
has
specified
in
the
final
rule
that
a
SSMP
is
required.
The
EPA
has
therefore
not
made
this
change
in
Table
8
of
the
final
rule.

The
EPA
has
noted
in
Table
8
of
the
final
rule
under
§
63.9(
b)(
1)­(
5)
that
§
63.9(
b)(
3)
is
reserved.
The
EPA
has
also
noted
in
Table
8
of
the
final
rule
under
§
63.9(
h)(
1)­(
6)
that
§
63.9(
h)(
4)
is
reserved.

A
SSMP
is
required
under
the
final
rule.
The
EPA
has
therefore
not
made
the
change
to
§
63.10(
b)(
2)(
i)­(
v)
that
the
commenter
suggested.
All
sections
of
63.10(
c)
are
applicable
since
EPA
is
requiring
a
SSMP.
The
EPA
has
noted
in
Table
8
of
the
final
rule
that
§
63.10(
c)(
2)­(
4)
112
and
(
9)
are
reserved.
The
EPA
is
requiring
immediate
SSM
reports
only
when
actions
addressing
the
startup,
shutdown,
or
malfunction
were
inconsistent
with
the
SSMP.
The
EPA
has
made
this
notation
in
Table
8
of
the
final
rule
under
§
63.10(
d)(
5).
The
EPA
assumes
the
commenter
means
that
§
63.10(
e)(
3)(
i)(
C)
is
a
reserved
paragraph
and
not
§
63.10(
3)(
i)(
C).
The
EPA
has
noted
in
Table
8
of
the
final
rule
that
§
63.10(
e)(
3)(
i)(
C)
is
a
reserved
paragraph.
The
EPA
disagrees
with
the
commenter's
request
of
a
No
response
to
§
63.10(
e)(
3).
The
EPA
has
not
made
this
change.

B.
Other
X.
B.
1
Comment:
Five
commenters
(
542,
551,
554,
556,
587)
asserted
that
the
basis
for
any
size
threshold
should
be
expressed
in
site­
rated
HP
as
opposed
to
manufacturer's
nameplate
HP.
One
commenter
(
587)
gave
the
following
reasons:

(
1)
The
database
used
by
EPA
to
determine
the
MACT
floor
provisions
likely
includes
the
site­
rated
HP,
based
on
the
facility's
air
permit;

(
2)
Stationary
RICE
are
typically
identified
by
site­
rated
HP,
rather
than
manufacturer's
nameplate
HP
in
the
facility's
title
V
permit
and
not
all
engines
have
HP
on
the
nameplate;
and
(
3)
Federal
Energy
Regulatory
Commission
certified
HP
for
natural
gas
transmission
facilities
are
issued
based
on
site­
rated
HP.

Response:
The
EPA
contacted
one
of
the
commenters
who
submitted
this
comment
and
also
an
engine
manufacturer.
Information
received
from
both
sources
indicated
that
there
may
be
differences
between
site­
rated
HP
and
the
manufacturer's
nameplate
rating.
Factors
such
as
altitude,
temperature,
fuel,
etc.
affect
what
the
site­
rated
HP
will
be
for
the
engine
at
a
specific
location.
Some
manufacturers
include
the
specific
site
rating
on
the
nameplate
of
the
engine,
which
is
a
HP
rating
which
has
been
adjusted
to
account
for
the
characteristics
of
the
location
the
engine
is
installed
at
as
well
as
other
parameters
affecting
the
engine
rating.
For
these
reasons,
EPA
agrees
with
the
commenter
that
it
is
appropriate
to
use
the
site­
rated
HP
as
opposed
to
the
manufacturer's
nameplate
rating
for
the
size
applicability
criteria,
because
relying
on
the
manufacturer's
nameplate
rating
may
not
be
representative
of
the
capability
of
the
engine
on­
site.
The
EPA
has
specified
this
in
the
final
rule.

X.
B.
2
Comment:
Four
commenters
(
542,
554,
556,
587)
asked
that
EPA
include
nonaggregation
provisions
for
transmission
and
storage
facilities
for
the
Transmission
&
Storage
(
T&
S)
MACT.

Response:
The
EPA
has
incorporated
this
comment
in
the
final
rule.
The
nonaggregation
provisions
for
transmission
and
storage
facilities
from
the
Natural
Gas
Transmission
and
Storage
MACT
(
40
CFR
part
63,
subpart
HHH),
which
are
found
in
the
definition
of
major
source
in
that
subpart,
are
as
follows:
113
(
1)
Emissions
from
any
pipeline
compressor
station
or
pump
station
shall
not
be
aggregated
with
emissions
from
other
similar
units,
whether
or
not
such
units
are
in
a
contiguous
area
or
under
common
control;
and
(
2)
Emissions
from
processes,
operations,
and
equipment
that
are
not
part
of
the
same
natural
gas
transmission
and
storage
facility,
as
defined
in
this
section,
shall
not
be
aggregated.

The
non­
aggregation
provisions
in
(
1)
above
were
already
included
in
the
proposed
definition
of
major
source
for
the
RICE
NESHAP
and
have
been
retained
in
the
final
rule.
The
non­
aggregation
provisions
in
(
2)
above
have
also
been
added
to
the
definition
of
major
source
for
the
RICE
NESHAP.

X.
B.
3
Comment:
Five
commenters
(
542,
549,
554,
556,
587)
requested
that
EPA
include
the
provisions
to
calculate
potential
emissions
for
storage
facilities
from
the
T&
S
MACT.

Response:
The
EPA
agrees
with
the
commenters
and
has
incorporated
their
comment
in
the
final
rule
by
modifying
the
definition
of
potential
to
emit
in
the
final
rule
to
include
the
following:

"
For
oil
and
natural
gas
production
facilities
subject
to
subpart
HH
of
this
part,
the
potential
to
emit
provisions
in
§
63.760(
a)
may
be
used.
For
natural
gas
transmission
and
storage
facilities
subject
to
subpart
HHH
of
this
part,
the
maximum
annual
facility
gas
throughput
for
storage
facilities
may
be
determined
according
to
§
63.1270(
a)(
1)
and
the
maximum
annual
throughput
for
transmission
facilities
may
be
may
be
determined
according
to
§
63.1270(
a)(
2)."

X.
B.
4
Comment:
Two
commenters
(
582,
586)
asked
that
EPA
list
diesel
PM
as
a
HAP.
Commenter
582
stated
that
if
EPA
fails
to
act
on
its
own
initiative,
the
commenter
will
submit
a
formal
listing
petition
to
EPA.
Accordingly,
the
commenter
urged
EPA
to
require
PM
standards
for
all
stationary
diesel
RICE.
Reduction
of
diesel
PM
through
the
use
of
PM
traps
on
diesel
RICE
is
highly
cost­
effective
and
will
help
protect
public
health
from
the
harmful
effects
of
diesel
PM.
The
commenter
provided
extensive
details
on
the
health
effects
of
diesel
PM.
Commenter
586
recommended
including
diesel
particulate
matter
in
this
MACT
and
including
limits
and
control
measures.

One
commenter
(
546)
stated
that
because
the
rule
does
not
address
diesel
PM,
the
rule
requirements
are
not
sufficient
to
meet
the
risk
reduction
goals
of
the
Urban
Air
Toxic
Strategy.
The
commenter
stated
that
the
rule
is
not
health
protective
because
it
only
regulates
organic
gases
and
ignores
diesel
PM.
The
rule
should
allow
State
and
local
agencies
to
regulate
diesel
PM
as
a
more
health
protective
alternative
for
addressing
the
risk
from
diesel
exhaust.
The
commenter
stated
that
the
cancer
unit
risk
factor
for
diesel
PM
in
50
times
greater
than
that
for
formaldehyde.
Major
health
agencies
agree
that
adverse
human
health
effects
result
from
environmental
exposure
to
diesel
PM
exhaust.
From
a
public
health
perspective,
focusing
on
diesel
PM
as
a
surrogate
for
whole
diesel
exhaust
is
a
more
protective
approach
compared
to
focusing
only
on
organics
as
114
proposed.

Response:
The
EPA
acknowledges
the
comments
on
this
issue.
However,
EPA
is
not
prepared
at
this
time
to
list
diesel
PM
as
a
regulated
HAP,
at
least
not
in
the
context
of
the
final
rule.
The
EPA
proposed
the
rule
for
the
purposes
of
promulgating
regulations
for
emissions
from
stationary
RICE
that
were
already
listed
under
section
112
of
the
CAA.
While
EPA
did
mention
the
diesel
exhaust
issue,
EPA
did
not
include
any
detailed
discussion
on
the
separate
issue
of
whether
any
additional
pollutants
should
be
added
to
the
list
of
regulated
pollutants
under
CAA
section
112.
The
decision
regarding
whether
to
list
diesel
PM
entails
several
significant
issues
that
have
not
been
discussed
in
the
context
of
the
final
rule.
This
includes
the
issues
of
what
adverse
health
effects
are
related
to
diesel
PM,
what
components
of
diesel
exhaust
should
be
regulated,
whether
the
regulation
of
particulate
matter
as
a
criteria
pollutant
is
relevant,
and
other
issues.
Therefore,
it
would
be
inappropriate
to
take
final
action
on
this
comment
in
the
context
of
the
final
rule.
The
EPA
discussed
DPF
which
can
reduce
PM
in
response
to
comment
IV.
C.
1.
The
EPA
will
be
addressing
the
Urban
Air
Toxics
strategy
implications
of
RICE
HAP
emissions
in
a
separate
rule
scheduled
to
be
completed
in
2007.

X.
B.
5
Comment:
One
commenter
(
579)
urged
EPA
to
go
through
rule
language
and
streamline
the
requirements
wherever
possible
to
reduce
the
burden
on
both
the
equipment
operators
and
the
regulatory
personnel
including
EPA
who
must
process
many
documents
to
implement
the
proposed
rule.
At
the
minimum,
many
of
the
requirements
should
be
re­
cast
into
flow
charts
so
that
all
the
requirements
for
each
type
of
engine
are
shown
sequentially.

Response:
The
EPA
has
tried
to
streamline
the
requirements
in
the
final
rule
as
best
as
possible.
However,
EPA
has
not
included
flow
charts
in
the
final
rule.
The
EPA
feels
that
the
regulatory
text
and
tables
sufficiently
indicate
the
requirements
of
the
final
rule.
Displaying
all
the
requirements
for
each
type
of
engine
separately
is
repetitive
and
not
necessary.

X.
B.
6
Comment:
One
commenter
(
564)
stated
that
the
Agency
should
further
review
the
rule
to
reduce
the
cost
and
regulatory
burden
aspects
of
the
rule
that
provide
no
significant
environmental
benefit.
With
all
the
reports
that
are
sent
in,
a
question
by
the
commenter
was
"
are
these
reports
regularly
reviewed
by
regulatory
Agencies?"
If
not,
and
the
commenter's
experience
suggested
they
are
not,
then
reporting
should
be
simplified.
Some
considerations
to
reduce
the
ongoing
cost
of
the
RICE
MACT:
Develop
a
regulatory
option
to
focus
on
the
facilities
that
pose
significant
risks
and
avoid
facilities
that
pose
little
risk.
Eliminate
the
operation
limitations
that
relate
to
catalyst
pressure
drop
and
temperature
differential.

Response:
The
EPA
feels
it
has
done
what
it
can
to
reduce
the
cost
and
regulatory
burden
aspects
of
the
rule
that
provide
no
significant
environmental
benefit.

X.
B.
7
Comment:
Two
commenters
(
526,
579)
expressed
concern
with
the
overall
extended
timeframe
of
the
approval
process
for
performance
tests
and
other
procedures
petition
requests
(
67
FR
77862).
The
timeframes
are
compounded
by
the
title
V
environment
of
public
review
and
comment
and
EPA
review
of
significant
permit
revisions.
This
in
turn
is
further
115
exacerbated
by
the
pre­
approval
requirement
at
§
63.6620(
f).
All
these
layered
requirements
should
be
re­
reviewed
by
EPA
comprehensively
and
streamlined
to
reduce
the
burden
or
regulated
entities
and
regulators.
The
existing
title
V
framework
including
periodic
monitoring
should
be
relied
upon
to
the
maximum
extent
possible
to
minimize
duplicate
and
redundant
requirements
under
this
proposed
MACT.

Response:
The
EPA
has
performed
a
thorough
review
of
the
requirements
and
feels
it
has
established
a
rule
that
is
appropriate
and
that
has
adequately
considered
the
impact
on
regulated
entities
and
regulators.

X.
B.
8
Comment:
One
commenter
(
565)
stated
that
recent
court
decisions
provide
EPA
with
significant
flexibility
when
promulgating
MACT
standards.
During
the
period
of
the
RICE
proposal
development,
the
D.
C.
Circuit
Court
of
Appeals
issued
two
decisions
­
National
Lime
and
CKRC
­
that
EPA
must
consider
when
establishing
NESHAP.
These
decisions
provide
significant
guidance
to
the
Agency
on
the
scope
of
its
rulemaking
and
the
methodology
that
EPA
must
consider
before
it
promulgates
a
final
rule.
While
placing
certain
restrictions
on
EPA's
authority
and
floor
determinations,
the
commenter
stated
that
EPA
still
retains
significant
amount
of
flexibility
in
setting
MACT
emission
limitations.

Response:
The
EPA
agrees
generally
with
the
comment
that
EPA
has
significant
flexibility
in
promulgating
MACT
standards.

X.
B.
9
Comment:
One
commenter
(
536)
stated
that
EPA
has
entered
into
a
proposed
settlement
with
the
Sierra
Club
regarding
those
MACT
standards
for
which
the
Agency
did
not
finalize
rules
by
the
statutory
MACT­
hammer
deadline.
The
EPA
consequently
proposed
a
twopart
extended
case­
by­
case
MACT
application
process
under
section
112(
j)
of
the
CAA,
with
part
1
due
May
15,
2002
and
part
2
due
May
15,
2004.
Accordingly,
the
Sierra
Club
challenged
the
2­
year
window
for
submission
of
part
2
case­
by­
case
MACT
applications.
The
current
settlement
with
Sierra
Club
provides
for
a
revised
part
2
application
deadline
for
RICE
MACT
of
April
28,
2004.
The
commenter
stated
that
submittal
of
part
2
case­
by­
case
MACT
applications
will
be
overly
burdensome
for
both
affected
sources
and
State
regulatory
agencies
as
well
as
could
result
in
inconsistent
MACT
determinations
to
individual
applicants.
This
part
2
deadline
would
only
apply
if
EPA
misses
the
deadline
for
final
MACT
promulgation
60
days
before
April
28,
2004.
Therefore,
the
commenter
urged
the
EPA
to
ensure
timely
finalization
on
the
proposed
RICE
MACT
standards
prior
to
the
relevant
settlement
trigger
date.

Response:
The
EPA
also
wants
to
see
timely
finalization
of
the
rule.
The
EPA
has
done
everything
possible
to
ensure
that
the
final
rule
is
completed
on
time,
which
will
result
in
part
2
applications
not
being
required.

X.
B.
10
Comment:
One
commenter
(
594)
stated
that
the
Colorado
Air
Quality
Control
Commission,
the
policy
and
rulemaking
entity
in
Colorado,
looked
at
engines
in
1993
and
again
in
2002
and
has
established
a
comprehensive
scheme
for
threshold
reporting
of
emissions
of
criteria
pollutants
and
HAP
on
air
pollutant
emission
notices
(
APEN)
and
threshold
levels
for
permit
for
116
criteria
pollutants.
Colorado's
Regulation
No.
3
(
5
CCR
1001­
5)
Part
A.
Section
II.
sets
forth
the
APEN
requirements.
Air
pollutant
emission
notices
reporting
requirements
for
HAP
emissions
are
based
on
the
relative
toxicity
of
the
substance
and
the
distance
of
the
emission
point
to
the
property
line
of
the
facility.
Appendix
A
of
Regulation
No.
3
sets
forth
various
HAP
reporting
thresholds
which
range
from
50­
250
lb
per
year
for
the
most
toxic
category
(
Bin
A)
and
1,000­
5,000
lb
per
year
for
the
least
toxic
(
Bin
C).
Formaldehyde
is
listed
in
Bin
A.
In
addition,
there
are
categorical
exemptions
from
APEN
reporting
requirements.

Response:
The
EPA
acknowledges
the
comments.

X.
B.
11
Comment:
One
commenter
(
579)
said
that
EPA
states
that
if
the
proposed
rule
is
implemented
at
all
affected
RICE
facilities,
annual
cancer
incidence
is
estimated
to
be
reduced
on
the
order
of
ten
cases
per
year.
The
commenter
suggested
that
this
number
be
clarified
or
eliminated,
especially
since
the
MACT
standards
are
not
risk­
based
and
EPA
cites
many
uncertainties
in
its
evaluation.

Response:
The
EPA
is
required
by
Executive
Order
12866
to
review
the
benefits
of
a
rule,
even
if
not
statutorily
required.
Moreover,
while
a
review
of
lives
saved
may
not
be
required
by
the
statute,
it
is
certainly
not
forbidden
for
EPA
to
review
the
expected
number
of
lives
saved
by
a
rule,
just
as
it
is
not
forbidden
to
review
the
expected
costs
of
a
rule,
which
is
not
a
factor
under
the
criteria
for
determining
the
MACT
floor.
In
any
event,
this
statement
was
included
in
the
proposal,
but
EPA
has
not
included
this
statement
in
the
final
rule.

X.
B.
12
Comment:
One
commenter
(
606)
asked
how
EPA
will
govern
this
rule
and
how
EPA
will
identify
existing
sites
that
have
rich
burn
engines
over
500
HP
that
require
NSCR
equipment.
The
commenter
further
asked
what
the
punishment
for
those
who
are
caught
not
complying
is.
The
commenter
also
asked
when
the
final
rule
will
begin.

Response:
Existing
4SRB
engines
will
have
3
years
from
the
date
of
publication
of
the
final
rule
in
the
Federal
Register
to
comply
with
the
final
rule.
The
RICE
final
rule
is
scheduled
to
be
published
in
March
or
April
of
2004.
That
means
that
existing
4SRB
engines
will
have
until
approximately
March
or
April
of
2007
to
comply.
After
the
rule
is
published
in
the
Federal
Register,
EPA
will
undertake
an
outreach
effort
to
inform
the
owners
and
operators
of
the
rule
and
its
requirements.
The
owners
and
operators
affected
by
the
rule
will
have
to
initially
notify
their
enforcing
agencies
within
a
certain
time
period.
The
EPA
will
use
industry
associations,
State
and
local
governments,
engine
manufacturers
and
control
device
vendors,
and
trade
magazines
to
notify
owners
and
operators
of
the
final
rule.
Any
owner
or
operator
who
has
not
notified
EPA
and
is
not
in
compliance
with
the
final
rule
at
the
end
of
the
3
year
period
will
be
in
non­
compliance
and
enforcement
action
will
be
taken.

X.
B.
13
Comment:
One
commenter
(
549)
stated
that
EPA
should
clarify
that,
pursuant
to
42
U.
S.
C.
§
7627,
the
Department
of
Interior
has
jurisdiction
over
the
central
and
western
Gulf
of
Mexico.
117
Response:
The
section
referred
to
by
the
commenter,
pertaining
to
air
pollution
from
outer
continental
shelf
activities,
has
been
the
subject
of
previous
EPA
rules.
The
EPA
refers
the
commenter
to
those
rules
regarding
the
application
of
that
provision.
