EPA-450/2-77-017a
September 1977
         STANDARDS SUPPORT
                  AND
 ENVIRONMENTAL IMPACT STATEMENT
  VOLUME 1:  PROPOSED STANDARDS
  OF PERFORMANCE FOR STATIONARY
             GAS TURBINES

       U.S. ENVIRONMENTAL PROTECTION AGENCY
           Office of Air and Waste Management
         Office of Air Quality Planning and Standards
        Research Triangle Park, North Carolina 27711

-------
                            EPA-450/2-77-017a
         STANDARDS SUPPORT
                  AND
ENVIRONMENTAL IMPACT STATEMENT
  VOLUME 1: PROPOSED STANDARDS
 OF PERFORMANCE FOR STATIONARY
             GAS TURBINES
           Emission Standards and Engineering Division
          U.S. ENVIRONMENTAL PROTECTION AGENCY
             Office of Air and Waste Management
           Office of Air Quality Planning and Standards
           Research Triangle Park, North Carolina 27711

                 September 1977

-------
This report has been reviewed by the Emission Standards and Engineering Division, Office of Air Quality Planning
and  Standards, Office  of Air and Waste Management, Environmental Protection Agency, and approved for
publicati n.  Mention of company or product names does not constitute endorsement by EPA.  Copies are
available free of charge to Federal employees, current contractors and grantees, and non-profit organizations - as
supplies permit - from the Library Services Office, Environmental Protection Agency, Research Triangle Park,
North Carolina 27711; or may be obtained, for a fee, from the National Technical Information Service, 5285 Port
Royal Road, Springfield, Virginia 22161.
                                    Publication No. EPA-450/2-77-017a

-------
                                   Draft

                           Standards Support and
                      Environmental  Impact Statement

                          Stationary Gas Turbines

                      Type of Action:   Administrative

                                Prepared by
Director, Emission Standards and Engineering Division             (Date)
Environmental  Protection Agency
Research Triangle Park, North Carolina  27711

                                Approved by
 ssistant Administrator                                           (Date)
Office of Air and Waste Management
Environmental Protection Agency
401 M Street, S.W.
Washington, D. C.  20460
Additional copies may be obtained or reviewed at:

Public Information Center (PM-215)
Environmental Protection Agency
Washington, D. C.  20460

-------
                             TABLE OF CONTENTS

                                                                      Page

LIST OF TABLES	

LIST OF FIGURES 	

Chapter 1.  SUMMARY	1-1

        1.1  PROPOSED STANDARDS 	

        1.2  ENVIRONMENTAL/ECONOMIC IMPACT	

        1.3  INFLATIONARY IMPACT	

Chapter 2.  INTRODUCTION	2-1

        2.1  AUTHORITY FOR THE STANDARDS	2-1

        2.2  SELECTION OF CATEGORIES OF STATIONARY SOURCES	2-4

        2.3  PROCEDURE FOR DEVELOPMENT OF STANDARDS OF
             PERFORMANCE	2-5

        2.4  CONSIDERATION OF COSTS	2-8

        2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS 	 2-10

        2.6  IMPACT ON EXISTING STANDARDS 	 2-11

        2.7  REVISION OF STANDARDS OF PERFORMANCE 	 2-12

        2.8  REFERENCES	2-12

Chapter 3.  THE STATIONARY GAS TURBINE INDUSTRY AND PROCESS 	 3-1

        3.1  GENERAL	3-1

        3.2  PROCESSES AND THEIR EMISSIONS	3-36

        3.3  REFERENCES	3-111

Chapter 4.  EMISSION CONTROL TECHNIQUES 	 4-1

        4.1  PARTICULATE EMISSIONS	4-1

        4.2  VISIBLE EMISSIONS	4-2

        4.3  S02 EMISSIONS	4-3

        4.4  HYDROCARBON AND CARBON MONOXIDE EMISSIONS	4-7

-------
                      TABLE OF CONTENTS (continued)

                                                                    Page

        4.5  NOY EMISSIONS	4-23
               A

        4.6  REFE\:NCES	4-99

Chapter 5.  MODIFICATION AND RECONSTRUCTION 	 5-1

        5.1  40 CFR Part 60 PROVISIONS FOR MODIFICATION AND
             RECONSTRUCTION 	 5-2

        5.2  APPLICABILITY TO STATIONARY GAS TURBINE
             INSTALLATIONS	5-4

        5.3  REFERENCES	5-8

Chapter 6.  ENVIRONMENTAL IMPACT	6-1

        6.1  AIR POLLUTION IMPACT	6-1

        6.2  WATER POLLUTION IMPACT 	 6-39

        6.3  SOLID WASTE DISPOSAL IMPACT	6-40

        6.4  ENERGY IMPACT	6-41

        6.5  NOISE IMPACT	6-44

        6.6  OTHER ENVIRONMENTAL CONCERNS 	 6-44

        6.7  REFERENCES	6-47

Chapter 7.  ECONOMIC IMPACT 	  .... 7-1

        7.1  MODEL PLANT SELECTION	7-1

        7.2  COST OF ALTERNATIVE EMISSION CONTROL SYSTEMS  	 7-6

        7.3  COST ANALYSIS OF GAS TURBINE EMISSION CONTROL
             SYSTEMS	7-41

        7.4  ECONOMIC IMPACT	7-69

        7.5  REFERENCES	7-87

Chapter 8.  RATIONALE 	 8-1

        8.1  SELECTION OF SOURCE FOR CONTROL	8-1

        8.2  SELECTION OF POLLUTANTS	8-3
                                     11

-------
                       TABLE OF CONTENTS (continued)

                                                                   Page

        8.3  SELECTION OF AFFECTED FACILITIES	8-6

        8.4  SELECTION OF THE BEST SYSTEM OF EMISSION
             REDUCTION	•'	8-9

        8.5  SELECTION OF FORMAT FOR THE STANDARDS	8-1 7

        8.6  SELECTION OF EMISSION LIMIT 	 8-20

        8.7  MODIFICATION/RECONSTRUCTION 	 8-33

        8.8  SELECTION OF MONITORING REQUIREMENTS	8-34

        8.9  SELECTION OF PERFORMANCE TEST METHODS 	 8-35

Appendix A.  EVOLUTION OF THE SELECTION OF THE  BEST SYSTEM OF
             EMISSION REDUCTION	A-l

Appendix B.  INDEX OF ENVIRONMENTAL IMPACT CONSIDERATIONS	 B-l

Appendix C.  EMISSION TEST DATA SUMMARY	C-l

Appendix D.  EMISSION MEASUREMENT AND CONTINUOUS MONITORING. .  . . D-l

Appendix E.  AIR QUALITY ANALYSES	E-1

Appendix F.  ENFORCEMENT ASPECTS 	 F-l

Appendix G.  DETERMINATION OF NITROGEN OXIDES AND OXYGEN
             EMISSIONS FROM STATIONARY GAS TURBINES	6-1
                                    iii

-------
                                 List of Tables
                                                                           Page
Table 3-1   American Firms Marketing Stationary Gas Turbines	3-4
Table 3-2   Stationary Gas Turbine Firms Ranked by Total 1974 Orders. .  .  .3-6
Table 3-3   Rreakdow  of Total 1974 Stationary Gas Turbine Orders and
            Share of Market	3-8
Table 3-4   Breakdown of Firms'  1974 Orders for Stationary Gas Turbines .  .3-9
Table 3-5   U.S. Gas Turbine Shipments to Domestic Electric Utilities .  .  .3-20
Table 3-6   1974-1980 Market Projections	3-31
Table 3-7   1974-1985 Market Projections	3-32
Table 3-8A  Methodolooy for Market Projections	3-34
Table 3-8B  Emissions from Stationary Gas Turbines Produced by Six
            Manufacturers	3-46
Table 3-9   NOX Correlations for Gas Turbines	3-80
Table 3-10  Summary of State Reflations	3-108
Table 3-11  state and Local Emission Standards Applicable to Gas Turbines .3-109
Table 3-12  Effect of Existing State and Local Standards on Pollutant
            Emission Levels from a 33 MW Gas Turbine	3-110
Table 4-1   Summary of CO and HC Emissions for Best Configurations of
            Each Combustor Concept at Idle Conditions	4-20
Table 4-2   Known Turbine Installations with Water or Steam Injection
            for NOX Control	4-28
Table 4-3   Water Quality Specifications	4-29
Table 4-4   Comparison of Consumptive Water Utilization for Different
            Schemes of Generating 60 MW of Power	4-30
Table 4-5   The Effects of Water and Steam Injection on Power Augmentation
            and Gas Turbine Efficiency	4-38
Table 4-6   Summary of Exhaust Emissions and Percent Reduction from
            Current Production Engines	4-78
Table 4-7   Summary from Test Results for Best Configurations of Each
            Combustor Concept at Sea-Level Take-Off Conditions	4-83
Table 4-8   1973 Test Results from the Westinghouse/Engelhard Catalytic
            Combustor Evaluation	4-87
Table 4-9   1974 Test Results from the Westinghouse/Engelhard Catalytic
            Combustor Evaluation	4-87
Table 4-10  Data for Commercial and Synthetic Fuels with the Engelhard
            CATCOM DXA-111 Catalyst	4-89
Table 4-11  Percent Reduction of NOX Emissions Using Dry Control
            Techniques	4-91
Table 6-1   Stack Parameters	6-3
Table 6-2   Summary of Maximum Ground Level Concentrations for Garrett
            Airesearch Gas Turbine, Model GTC8S-90	6-11
Table 6-3   Summary of Maximum Ground Level Concentrations for Garrett
            Ai research Gas Turbine  (Cluster Arrangement)	6-12
Table 6-4   Summary of Maximum Ground Level Concentrations for Solar
            Gas Turbine, Model Saturn	6-13
Table 6-5   Summary of Maximum Ground Level Concentrations for Solar
            Gas Turbine  (Cluster Arrangement)	6-14
Table 6-6   Summary of Maximum Ground Level Concentrations for Solar
            Gas Turbine, Model Centaur	6-15
                                         iv

-------
                           List of Tables (continued)
                                                                           Page
Table 6-7   Summary of Maximum Ground Level Concentrations for Solar
            Gas Turbine (Cluster Arrangement)	6-16
Table 6-8   Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 5001 P	6-17
Table 6-9   Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-19
Table 6-10  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 5002 B	6-20
Table 6-11  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-21
Table 6-12  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 7901 B (Simple Cycle) .  .  . .   . 6-22
Table 6-13  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-23
Table 6-14  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 7001 B (Combined Cycle).  . .   . 6-25
Table 6-15  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-26
Table 6-16  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 7001 B (Regenerative Cycle) .   . 6-27
Table 6-17  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-28
Table 6-18  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine, Model MS 7001 C	6-30
Table 6-19  Summary of Maximum Ground Level Concentrations for General
            Electric Gas Turbine (Cluster Arrangement) 	 6-31
Table 6-20  Evaluation of Wind Velocity Increases on Maximum Ground
            Level Concentrations for Regenerative Cycle Gas Turbine.  . .   . 6-32
Table 6-21  Evaluation of Wind Velocity Increases on Maximum Ground
            Level Concentrations for 16-Unit Cluster of Gas Turbines  . .   . 6-33
Table 6-22  Evaluation of Wind Velocity Increases on Maximum Ground
            Level Concentrations for Spinning Reserve Mode Operation  . .   . 6-34
Table 6-23  Evaluation of Stack Height Increase on Maximum Ground
            Level Concentrations for Regenerative Cycle Turbine	6-37
Table 6-24  Evaluation of Stack Height increase on Maximum Ground
            Level Concentrations for Spinning Reserve Mode Operation  . .   . 6-38
Table 6-25  Additional Fuel Consumed in 1980 if All Turbines Sold in
            1974 Through 1980 Used Water Injection to Meet Regulations
            for NOX Emissions	6-42
Table 6-26  Estimated Impact of Three NOX Emission Levels for Gas Turbines
            When a Seven-Year Period is Considered 	 6-46
Table 7-1   Model Plant Characteristics	7-3
Table 7-2   Emission Rates for Non-Regenerated Turbines	7-7
Table 7-3   Water Quality Specifications 	 7-8
Table 7-4   Assumed Flow Rates for Alternative Standards 	 7-10
Table 7-5   Cost of Turbine Modifications	7-21
Table 7-6   Variation in Water Flow Requirements for 75 ppm Emission
            Level	7-23
Table 7-7   Variation in Water Flow Requirements for 125 ppm Emission
            Level	7-24

-------
                           List of Tables (continued)
                                                                        Page
Table 7-8   Summary of Model  Plant Costs for Water Injection for
            a 75 ppm Imission Limit	7-26
Table 7-9   Direct lesulfurization Costs	7-33
Table 7-10  Indirect Desulfurization Costs	7-34
Table 7-11  Mix of Crude Oils Used to Produce Residual  Oils 1n U.S.
            Markets	7-35
Table 7-12  Distribution of Residual Fuel  Oil Demand in 1980	7-36
Table 7-13  Distribution of Distillate Oil by Sulfur Content - 1974. .   .7-38
Table 7-14  Cost of Particulate Control  for Model  Plants	7-40
Table "M5  Natural Gas Intrastate Gas Prices	7-45
Table 7-16  Utility Operation arid Maintenance Experience with Gas
            Turbines	7-46
Table 7-17  Baseline Cost Estimates for Model Plant Applciations .  . .   .7-48
Table 7-18  Baseline Cost Sensitivity Analysis	7-51
Table 7-19  Baseline Cost Estimates Used to Compare the Incremental
            Cost of Environmental Controls	7-54
Table 7-20  Capital and Operating Cost Associated with Water Purifica-
            tion Equipment	7-56
Table 7-21  Estimated Average Cost per Thousand Gallons of Water 1n
            1974	*	7-58
Table 7-22  Parameter Estimates for Model  Plant Cost Evaluation	7-60
Table 7-23  Gas Turbine Environmental Control Costs	7-61
Table 7-24  Sensitivity Analysis of Environmental  Control Costs	7-65
Table 7-25  Impact of NOv Emission Control on the Installed Capital
            Cost of Gas Turbines	7-70
Table 7-26  Potential Impact of NO  Controls on Product Prices for the
            Seven Most Energy Intensive Industries	7-77
Table 7-27  Sales Projections and Incremental Capacity	7-79
Table 7-28  Total Capital and Annualized Cost	7-81
Table 7-29  Energy Penalty	7-82
Table 7-30  1974 Sales Volumes and Market Shares	7-85
Table 7-31  Cost-Effectiveness Comparison	7-86

-------
                            List of Figures
                                                                         Page
Figure 3-1   Cut-Away View of a Typical Stationary Gas Turbine	3-2
Figure 3-2   Yearly Domestic Orders for Stationary Gas Turbines by
             Application, Given in Units	3-14
Figure 3-3   Yearly Domestic Order for Stationary Gas Turbines by
             Application, Given in Horsepower	3-15
Figure 3-4   Cumulative Domestic Orders for Stationary Gas Turbines
             by Application	3-17
Figure 3-5   Typical Simple Cycle Gas Turbine	3-37
Figure 3-6   Cut-Away View of a Typical Simple Cycle Gas Turbine	3-38
Figure 3-7   Cut-Away View of a Simple Cycle, Two-Shaft Turbine with
             an Annular Combustor	3-40
Figure 3-8   Single Shaft Gas Turbine with an External Combustion
             Chamber	3-41
Figure 3-9   Typical Regenerative Cycle Gas Turbine	3-43
Figure 3-10  Typical Combined Cycle Gas Turbine	3-44
Figure 3-11  Can Type Combustion System for Gas Turbines	3-52
Figure 3-12  Visible Emissions vs. Load for a Turbine using Various
             Fuels	3-56
Figure 3-13  Visible Emissions vs. Load for Two Fuels with Different
             Hydrogen Contents	3-57
Figure 3-14  Visible Emissions vs. Fuel Hydrogen Content for Two
             Combustor Designs	3-57
Figure 3-15  Visible Emissions vs. Firing Temperature for a GE MS-5001
             Turbine Burning Residual Fuel	3-57
Figure 3-16  Visible Emissions vs. Shaft Horsepower for Various
             Compressor Inlet Temperatures	3-59
Figure 3-17  Hydrocarbon Emissions vs. Fuel-to-Air Ratio for Gas
             Turbines Burning Oil Fuel	3-61
Figure 3-18  Carbon Monoxide Emissions vs. Firing Temperatures for
             Gas Turbines Burning Oil Fuel	3-61
Figure 3-19  Relationships Between Combustion Efficiency and Levels
             of CO and HC	3-63
Figure 3-20  Relationships Between Idle Power Combustion Efficiency
             Levels of CO and HC Emissions, and Engine Cycle Pressure
             Ratios	3-64
Figure 3-21  Effect of Inlet Temperatyre on CO and HC Emissions	3-65
Figure 3-22  Effect of Inlet Pressure on CO and HC Emissions	3-65
Figure 3-23  CO Emissions vs. Turbine Load for Various Fuels	3-67
Figure 3-24  HC Emissions vs. Turbine Load for Various Fuels	3-67
Figure 3-25  Emissions vs. Load for a Turbine Burning Two Different
             Fuels	3-67
Figure 3-26  CO Emissions vs. Power Output for a Gas Turbine Burning
             Methanol and No. 2 Distillate Oil	\  .  .3-69
Figure 3-27  Effect of Heating Value on CO Emissions from Gas Turbine
             Combustors Burning Coal Gases of 105 to 200 Btu/scf	3-70
Figure 3-28  Effect of Heating Value on CO Emissions from Gas Turbine
             Combustors Burning Coal Gases of 90 to 105 Btu/scf	3-70
Figure 3-29  Thermochemical  Equilibrium Levels for NO  Emissions	3-74
Figure 3-30  Formation Rate Data for NOV	x	3-75
                                       A

-------
                      List of Figures (continued)
                                                                       Page
Figure 3-31  NO  Emissions vs.  Combustor Inlet Temperature as
             Presented by Lipfert	3-76
Figure 3-32  The Affect of Ambient Temperature and Humidity on NO
             Emissions from a Turbine	 3-77
Figure 3-33  Influence of Residence Time on NO  Emissions	3-77
Figure 3-34  Correction Factors for Ambient Humidity 	 3-82
Figure 3-35  NO  Emissions vs.  Ambient Temperature as Predicted by
             Various Correlations	3-83
Figure 3-36  NO  vs. Shaft Horsepower for Various Compressor Inlet
             Temperatures	3-84
Figure 3-37  NOX Yield Fraction vs. Fuel-Bound Nitrogen Content. .  .   . 3-88
Figure 3-38  NO  Emissions vs.  Load for a Turbine Burning Various
             Fuels	3-90
Figure 3-39  NO  Emissions vs.  Firing Temperature When Burning Crude
             and Distillate Fuels	3-91
Figure 3-40  NO  Emissions vs.  Power Output for a Turbine Burning
             No. 2 Oil and Methanol	3-92
Figure 3-41  Measured and Predicted NO  Emissions from Burning Various
             Fuels in a Gas Turbine Coal Gas Combustor	3-94
Figure 3-42  NO  Emissions from Full and Small Scale Combustors vs.
             Combustor Gas Exit Temperature When Burning Natural Gas
             and Coal Gas	3-95
Figure 3-43  Estimated NO  Emissions vs. Power Output for a Gas
             Turbine Burning Low Btu Gas, Natural Gas, and No. 2
             Distillate	3-97
Figure 3-44  Improvement in the Heat Rate of Simple Cycle Gas Turbine
             Since 1960	3-98
Figure 3-45  Simple Cycle and Combined Cycle Gas Turbine Heat Rate
             and Efficiencies for 1975 and Projections for 1980. .  .   . 3-99
Figure 3-46  Projected HHV Heat Rates for Simple Cycle Gas Turbines
             vs. Firing Temperature for Various Compression Ratios .   . 3-100
Figure 3-47  Heat Rate vs. Time of Order for Simple Cycle Gas
             Turbines	3-101
Figure 3-48  Heat Rate vs. Time of Order for Combined Cycle Generating
             Units	3-102
Figure 3-49  Gas Turbine Plant Performance for Combined, Regenerative
             and Simple Cycle Gas Turbines	3-103
Figure 3-50  NO  Emissions vs.  Firing Temperature for a Simple Cycle
             Turbine Burning Distillate Fuel  	 3-106
Figure 3-51  NO  Emissions vs.  Firing Temperature for a Regenerative
             Cycle Turbine Burning Distillate Fuel  	 3-107
Figure 4-1   Comparison of Gas Turbine Smoke Emission Characteristics. 4-4
Figure 4-2   Effect of Fuel Additives and Combustor Redesign on
             Visible Emissions 	 4-5
Figure 4-3   S0? Emission Rate as a Function of Fuel Sulfur Content
             ana Gas Turbine Fuel Flowrate	4-6
Figure 4-4   CO Emissions vs. Turbine Size for Small Gas Turbines
             Without NO  Controls	4-9
Figure 4-5   CO Emissions vs. Turbine Size for Large Gas Turbine
             Without NOV Controls	4-10
                       A

-------
                      List of Figures (continued)
Figure 4-6

Figure 4-7A

Figure 4-7B


Figure 4-8

Figure 4-9

Figure 4-10

Figure 4-11

Figure 4-12

Figure 4-13

Figure 4-14

Figure 4-15
Figure 4-16

Figure 4-17

Figure 4-18

Figure 4-19

Figure 4-20

Figure 4-21

Figure 4-22
Figure 4-23

Figure 4-24

Figure 4-25
Figure 4-26

Figure 4-27

Figure 4-28
Figure 4-29
Figure 4-30
Figure 4-31
HC Emissions vs. Turbine Size for Small Gas Turbines
Without NO  Controls	
HC Emissions vs. Turbine Size for Large Gas Turbines
Without NO  Controls	
CO and HC Emission Reductions with Improved Fuel
Atomization in an Aircraft Gas Turbine at Ground
Idle Power Operating Conditions 	
CO and HC Emission Reductions Using Various Concepts of
Staged Fuel Injection 	
CO and HC Emission Reductions vs. Compressor Bleed Air
Extraction	
CO and HC Emission Reductions Using Various Combustor
Design Modifications	
CO Emission Characteristics for the General Electric
CF6-50 Production Engine/Combustor	
HC Emission Characteristics for the General Electric
CF6-50 Production Engine/Combustor	
Summary of NO  Emission Data from Gas Turbines Using
Wet Control Techniques	
Effectiveness of Water/Steam Injection in Reducing NOX
Emissions 	
Typical Water Purification System for Five Gas Turbines  .
Predicted NO  Emissions for a Range of Nitrogen Contents
and Water Injection Rates 	
Calculated NO  Reduction as a Function of Water Injection
Rate	
Calculated NO  Reduction as a Function of Steam Injection
Rate	x	
Factors Effecting the Pollutants from Gas Turbines at
Low and High Powers
CO and NO  Performance of Conventional Gas Turbine
Combustorx	
Effect of Variable Geometry Control on Gas Turbine
Emissions 	
CO vs. NO  for Rig Tests of the T-63 Combustor	
NO  Emission Reductions by Combustor Design Modification
on Aircraft Gas Turbine 	
Effect of Primary Zone Equivalence Ratio on NO  Emissions
from an Aircraft Gas Turbine Combustor	
NO  Emissions vs. Combustor Temperature Rise	
Effect of Exhaust Gas Recirculation on NO Emissions
When Burning No. 2 Distillate Fuel	
Effect of Exhaust Gas Recirculation on NO Emissions
When Burning Natural Gas	
Details of the Vortex Air Blast (VAB) Combustor 	
VAB Combustor - Fuel Injection Modification 	
VAB Combustor - Reaction Zone Diameter Increase 	
VAB Combustor - Swirler Throat Length Increase	
Page


4-11

4-12


4-15

4-16

4-17

4-18

4-21

4-22

4-25

4-27
4-32

4-37

4-40

4-41

4-43

4-44

4-46
4-47

4-49

4-51
4-52

4-53

4-54
4-56
4-57
4-58
4-59
                                     IX

-------
                      List of Figures (continued)

                                                                      Page

Figure 4-32  VAB Combustor - Reaction Zone Length Increase	4-60
Figure 4-33  VAB F-iel Injection Modification	4-61
Figure 4-34  VAB Lombustor NO  Test Results	4-62
Figure 4-35  Basic Jet-Induced Circulation (JIC) Combustion
             Details	4-63
Figure 4-36  JIC Combustor Modifications	4-64
Figure 4-37  Results from Rig Test of the JIC Combustor	4-65
Figure 4-38  Production CF6-50 Combustor Cross-section	4-68
Figure 4-39  Production CF6-50 Combustor Assembly 	 4-69
Figu/e 4-40  General Arrangement, Singular Annualar Combustor .... 4-70
Figure 4-41  Lean Dome Double Annular Combustor for CF6-50 Engine .   . 4-71
Figure 4-42  General Arrangement, Double Annular Dome Combustor . .   . 4-72
Figure 4-43  General Arrangement, Radial/Axial Staged Combustor . .   . 4-73
Figure 4-44  Radial/Axial Staged Combustor Assembly 	 4-74
Figure 4-45  Baseline Swirl-Can-Modular Combustor Design	4-76
Figure 4-46  NO  Emission Levels, Best Configuration of Each Major
             Design Approach	4-77
Figure 4-47  Swirl-Can Combustor Concept	4-80
Figure 4-48  Staged Premix Combustor Concept	4-81
Figure 4-49  Swirl Vorbix Combustor Concept 	 4-82
Figure 4-50  Catalytic Combustor Design Concept 	 4-85
Figure 4-51  Temperature History for Conventional Combustors, Lean
             Pre-Mix Combustors, and Catalytically Supported
             Combustion	4-86
Figure 4-52  Summary of NO  Emission Data from Gas Turbines Using
             Dry Control Techniques  	 4-92
Figure 4-53  Summary of NO  Emission Data from Gas Turbines Using
             Both Dry and Wet Controls	4-93
Figure 4-54  NO  Emissions as a Function of the Type of Fuel Used .  . 4-95
Figure 6-1   Mociel Plant Cluster Arrangements	6-5
Figure 7-1   Water Transport Costs	7-16
Figure 7-2   Comparative Cost for Gas Turbines and Diesel
             Generators	7-72
Figure 8-1   Summary of  NOX Emission Data from Gas Turbines Using
             Wet Control Techniques  	 8*21
Figure 8-2   Relationship Between Efficiency  and Uncontrolled
             NOX Emissions	8-23
Fiqure 8-3   Variation in Fuel-Bound Nitrogen Content of  Petroleum
             Fuels	8-26
Figure 8-4   Variation in Conversion of Fuel-Bound Nitrogen
             to Organic  NOX	8r27
Figure 8-5   Controlled  NO  Emission Levels vs.  Fuel-Bound Nitrogen
             Level.	8-28

-------
                              1.   SUMMARY
1.1  PROPOSED STANDARDS
     Standards of performance for stationary gas turbines are being proposed
under section 111 of the Clean Air Act.   These standards would limit emissions
of nitrogen oxides and sulfur dioxide from stationary gas turbines whose
peak load is equal to or greater than 10.7 gigajoules per hour heat input.
     The numerical emission limit for NOX would be 75 ppm by volume corrected
to 15 percent oxygen and ISO ambient atmosphere conditions.   The proposed
standard would also include an adjustment factor for gas turbine efficiency
and a fuel-bound nitrogen allowance.  NOx emissions from gas turbines, there-
fore, would be limited according to the  following equation:

                         STD = (0.0075 E) + F
     where:
     STD = allowable NOX emissions (percent by volume at 15  percent oxygen)
     E   = efficiency adjustment factor:
           14.4 kilgjgules/watt-hr
            Actual ISO heat rate
     F   = fuel-bound nitrogen allowance:
           Fuel-Bound Nitrogen                            F
           (percent by weight)                (NOX - percent by volume)
                N < 0.015                                 0
             0.015 < N < 0.1                           0.04(N)
             0.1 < N < 0.25                     0.004 + 0.0067 (N-0.1)
                N > 0.25                                0.005
                                    1-1

-------
During performance tests to determine compliance with  the  proposed standard,
measured NOX emissions at 15 percent oxygen would be adjusted to ISO ambient
atmospheric cond-"-.ions by the following correction factor:
                              p
               NOX = (NOX u XTT^O.B e!9(Hobs-0.00633)
                        Aobs  robs
     where:
     NOX    = emissions of NOX at 15 percent oxygen and ISO standard
              ambient conditions.
     NOX ,   = measured NOX emissions at 15 percent oxygen, ppmv.
     Pref   = reference combustor inlet absolute pressure at 101.3
              kilopascals  (1 atmosphere) ambient pressure.
     pobs   = nieasured combustor inlet absolute pressure.
     H^   = specific humidity of ambient air.
     e      = transcendental constant  (2.718).
     Gas turbine manufacturers, owners, or operators may develop their own
factors for use during performance tests to adjust measured NOX emissions
to  ISO ambient atmospheric conditions.  These factors, however, must be
in  terms of the following  variables:   combustor inlet pressure, ambient air
pressure,  ambient air humidity and ambient air temperature.  These factors
must also  be substantiated with data and approved by the Administrator for
use.
     The proposed standard for NO/ emissions would not apply to stationary
gas turbines with a peak load heat input of less than 107.2 gigajoules per
hour until five years following proposal.  This would provide time for manu-
facturers  to incorporate NOX controls  on these turbines.
     Emergency gas turbines, fire-fighting gas turbines, military gas turbines
and gas turbines creating  ice fog would be exempt from the NOX standard.
                                  1-2

-------
     The numerical emission limit for $03 would be 150 ppm by volume corrected
to 15 percent oxygen or a fuel  sulfur content limit of 0.8 percent by weight.
There would be no efficiency adjustment factor or ambient condition correction
factor for S02 emissions, since S02 emissions are not affected by gas turbine
efficiency or ambient atmospheric conditions.  Also, no-gas turbines would
be exempt from the S02 emission limit or fuel sulfur content limit except
those gas turbines whose peak load is less than 10.7 gigajoules per hour
heat input.
     The proposed standard requires continuous monitoring of the water-to-
fuel ratio of a turbine using water or steam injection to control NOX
emissions and daily determination of the sulfur and nitrogen contents of
the fuel fired by the gas turbine.

1.2  ENVIRONMENTAL/ECONOMIC IMPACT
     Two alternative emission control systems were considered for selection
as the basis for standards of performance limiting NOX emissions from
stationary gas turbines.  These alternatives were dry controls, which
consist mainly of different gas turbine combustor designs; and wet controls,
which consist of water or steam injection into the gas turbine combustor.
Dry controls would require approximately five years to incorporate on
most gas turbines, whereas wet controls could be applied to large gas
turbines now and could be applied to small gas turbines 1n about three years.
     The main environmental benefit of a standard based on wet controls
would be a reduction in NOX emissions, totaling about 190,000 tons per year
in 1981.  Adverse environmental impacts associated with wet controls would
be negligible.  The impact on water and solid waste pollution would be
small because the wastewater can be sewered directly.  Energy consumption
                                  1-3

-------
in the United States would increase only about 0.03 percent in 1981  due
to wet controls on large gas turbines.   There would be another 0.03  percent
increase in energy consumption in 1986  due to application of wet controls
on small gas turbines.
     Standards based on dry controls would have no impact on national NOV
                                                                        A
emissions in 1981 due to the necessity  of allowing a five-year delay to
incorporate dry controls on gas turbines.  Dry controls on gas turbines
would reduce NOX emissions by about 90,000 tons per year in 1986.  There
would, however, be no water pollution,  solid waste pollution, noise  pollution
or increased energy consumption associated with dry controls.
     The economic impact associated with standards limiting NO  emissions
                                                              A
based on wet or dry controls would be approximately the same and would be
small.  The total annualized cost associated with the use of wet controls
or dry controls would translate into price increases for the end products
or services provided by gas turbine users of from 0.01 percent in the
petroleum refining industry to 0.1 percent in the electric utility industry.
     The environmental and economic impacts associated with these two control
alternatives and the alternative of having no standard to limit NOX emissions
are summarized in Table 1.  Based on this assessment of the impacts of
standards of performance based on wet controls or dry controls, wet controls
are selected as  "...the best system of emission reduction (considering costs)..."
for the reduction of NOV emissions.
                       A
     There  are two possible control techniques for reducing $02 emissions
from stationary  gas turbines:  flue gas desulfurization  (FGD) and the firing
of low sulfur fuels.  FGD would cost about two to three times as much as
the gas turbine.  The economic impact of standards of performance based on
FGD, therefore,  is not considered reasonable.

                                  1-4

-------
     Nearly all  gas turbines currently fire low sulfur fuels and this
situation Is expected to continue at least through 1985.   The Impact on
ambient air quality of standards of performance based on  the firing of
low sulfur premium distillate fuel oils, therefore, would be negligible.
The economic impact would also be negligible and there would be no water,
energy, solid waste or noise impacts associated with standards based on
the firing of low sulfur fuel oils.
     Based on this assessment of the impacts of standards of performance
based on the firing of low sulfur fuel oils, this control technique is
selected as "...the best system of emission reduction (considering costs)..."
for the reduction of S02 emissions.
1.3  ECONOMIC IMPACT
     An Economic Impact Analysis would be developed if the proposed standard
caused an increase in operating costs in the fifth year of more than
$100 MM, a major product price increase of 5 percent, or an Increase in
national energy consumption of 25,000 barrels of fuel oil per day.  The
proposed standard for stationary gas turbines would Increase annuallzed
costs by about 11 million do-llars for large turbines by 1981 and by about
30 million dollars for all turbines by 1986.  Energy consumption would
Increase by 5500 barrels of fuel oil per day for large turbines by 1981
and 7000 barrels per day for small turbines by 1986.  The proposed standards
would result in a price increase of 7.7 percent for gas turbines used on
offshore platforms; however, a vast majority of these offshore turbines
will have a peak load of less than 107.2 gigajoules per hour and will
thus be exempt from the standard for five years.  Some of the larger new
offshore platforms will require turbines of peak load greater than 107.2
                                  1-5

-------
gigajoules per hour; however, these turbines will  constitute such a small
percentage of the overall market that the net price increase over the
entire turbine m--ket will be about 1 to 4 percent.  The Agency, therefore,
feels that no Economic Impact Analysis is required.
                                  1-6

-------
 ro
 C
 to
 o>
 ro
 c
 S-
 0)
                                 C
                                 o
                                 ro

                                t
                                 C
                                 O


                                 o
                                 c
                                 o
                                 u
                                   .
                                 CT
                                 s-
                                 O)
                                 O)

                                •I—
                                 o
 o
 (O
 Q.
 E
 u
•i—

 O
 c
 o
 o
UJ

TD
 c
 re
 fO
 

-M CD  i-
 C t/1  ro
 O O T3
C_- Q. C
    o  
 Q} n  (/;
-tr- s.	,
                                                                       o
                                                                     Q.     O
                                                                     E      <0
                                                                     t— i  4->  O  4->
                                                                         U  E  O
                                                                       i— i
                                                                  Q- C        0.'
                                                                                   C  >
                                                                                   O) -O
                                                              1-7

-------
                           2.   INTRODUCTION

     Standards of performance under section 111  of the Clean Air Act are
proposed following a detailed investigation of air pollution control
methods available to the affected industry and the impact of their costs
on the industry.  This document summarizes the information obtained from
such a study of gas turbines.   Its purpose is to explain in detail the
background and basis of the proposed standards and to facilitate analysis
of the proposed standards by interested persons, including those who may
not be familiar with the many technical aspects of the industry.  To
obtain additional copies of this document or the Federal Register notice
of proposed standards, write to Public Information Center (PM-215),
Environmental Protection Agency, Washington, D.C. 20460 (specify "Standards
Support and Environmental Impact Statement - An Investigation of the Best
Systems of Emission Reduction for Stationary Gas Turbines").
2.1  AUTHORITY FOR THE STANDARDS
     Standards of performance for new stationary sources are developed
under section 111 of the Clean Air Act (42 U.S.C. 1857c-6), as amended
in 1970.  Section 111 requires the establishment of standards of performance
for new stationary sources of air pollution which "...may contribute
significantly to air pollution which causes or contributes to the endanger-
ment of public health or welfare."  The Act requires that standards of
performance for such sources reflect "...the degree of emission limitation
achievable through the application of the best system of emission reduction
                                  2-1

-------
which (taking into account the cost of achieving such reduction) the
Administrator determines has been adequately demonstrated."  The standards
apply only to stationary sources, the construction or modification of
which commences ufter regulations are proposed by publication in the
Federal Register.
     Section 111 prescribes three steps to follow in establishing standards
of performance.
     1.  The Administrator must identify those categories of stationary
         sources for which standards of performance will ultimately be
         promulgated by listing them in the Federal Register.
     2.  The regulations applicable to a category so listed must be
         proposed by publication in the Federal Register within 120 days
         of its listing.  This proposal provides interested persons an
         opportunity for comment.
     3.  Within 90 days after the proposal, the Administrator must
         promulgate standards with any alterations he deems appropriate.
     Standards of performance, by themselves, do not guarantee protection
of health or welfare; that is, they are not designed to achieve any
specific air quality levels.  Rather, they are designed to reflect best
demonstrated technology (taking into account costs) for the affected
sources.  The overriding purpose of the collective body of standards is
to maintain existing air quality and to prevent new pollution problems
from developing.
     Previous  legal challenges to standards of performance have resulted
                           1 2
in several court decisions  '  of importance in developing future standards.
In cases the principal  issues were whether EPA:   (1) made reasoned
decisions and  fully explained the basis of the standards,  (2) made
available to interested parties the information on which the standards were
                                   2-2

-------
based, and (3) adequately considered significant comments from interested
parties.
     Among other things, the court decisions established:  (1) that
preparation of environmental impact statements is not necessary for
standards developed under section 111 of the Clean Air Act because,
under that section, EPA must consider any counter-productive environmental
effects of a standard in determining what system of control is "best;"
(2) in considering costs it is not necessary to provide a cost-benefit
analysis; (3) EPA is not required to justify standards that require
different levels of control in different industries unless such different
standards may be unfairly discriminatory; and (4) it is sufficient for
EPA to show that a standard can be achieved rather than that it has been
achieved by existing sources.
     Promulgation of standards of performance does not prevent State or
local agencies from adopting more stringent emission limitations for the
same sources.  On the contrary, section 116 of the Act (42 U.S.C. 1857-
D-l) makes clear that States and other political subdivisions may enact
more restrictive standards.  Furthermore, for heavily polluted areas,
more stringent standards may be required under section 110 of the Act
(42 U.S.C. 1857c-5) in order to attain or maintain national ambient air
quality standards prescribed under section 109 (42 U.S.C. 1857c-4).
Finally,  section 116 makes clear that a State may not adopt or enforce
less stringent new source standards than those adopted by EPA under
section 111.
     Although standards of performance are normally structured in terms
of numerical emission limits where feasible,-  alternative approaches are
sometimes necessary.  In some cases physical measurement of emissions
from a new source may be impractical or exorbitantly expensive.   For
                                    2-3

-------
example, emissions of hydrocarbons from storage vessels for petroleum
liquids are greatest during storage and tank filling.   The nature of the
emissions (high concentrations for short periods during filling and low
concentrations for longer periods during storage) and the configuration
of storage tanks make direct emission measurement impractical.  Therefore,
a more practical approach to standards of performance for storage vessels
has been equipment specifications.
 2.2  SELECTION OF CATEGORIES OF STATIONARY SOURCES
     Section 111 directs the Administrator to publish and from time to
time revise a list of categories of sources for which standards of
performance are to be proposed.  A category is to be selected "...if
[the Administrator] determines it may contribute significantly to air
pollution which causes or contributes to the endangerment of public
health or welfare."
     Considerable attention has been given to the development of a
system for assigning priorities to various source categories.  In brief,
the approach that has evolved is as follows.  Specific areas of interest
are identified by considering the broad strategy of the Agency for
implementing the Clean Air Act.  Often, these "areas" are actually
pollutants which are primarily emitted by stationary sources.  Source
categories which emit these pollutants are then evaluated and ranked by
a process involving such factors as  (1) the level of emission control
(if any) already required by State regulations;  (2) estimated levels
of control that might result from standards of performance for the
source category; (3) projections of growth and replacement of existing
facilities for the source category; and (4) the estimated incremental
amount of air pollution that could be prevented, in a pre-selected
                                  2-4

-------
future year, by standards of performance for the source category.  An
estimate is then made of the time required to develop a standard.  In
some cases, it may not be feasible to develop a standard immediately for
a source category with a high priority.  This might occur because a
program of research and development is needed to develop control techniques
or because techniques for sampling and measuring emissions may require
refinement.
     Selection of the source category leads to another major decision:
determination of the types of facilities within the source category to
which the standard will apply.  A source category often has several
facilities that cause air pollution.  Emissions from some of these
facilities may be insignificant or very expensive to control.  An investigation
of economics may show that, within the costs that an owner could reasonably
afford, air pollution control is better served by applying standards to
the more severe pollution problems.  For this reason (or perhaps because
there may be no adequately demonstrated system for controlling emissions
from certain facilities), standards often do not apply to all sources
within a category.  For similar reasons, the standards may not apply to
all air pollutants emitted by such sources.  Consequently, although a
source category may be selected to be covered by a standard of performance,
not all pollutants or facilities within that source category may be
covered by the standards.
2.3  PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
     Congress mandated that sources regulated under section 111 of the
Clean Air Act be required to utilize the best system of air pollution
control (considering costs) that has been adequately demonstrated at the
time of their design and construction.  In so doing, Congress sought to:
                                    2-5

-------
     1.   Maintain existing high-quality air,
     2.   Prevent new air pollution problems, and
     3.   Ensure uniform national  standards for new facilities.
     Standards or performance, therefore, must (1) realistically reflect
best demonstrated control practice; (2) adequately consider the cost of
such control; (3) be applicable to existing sources that are modified as
well as  new installations; and (4) meet these conditions for all variations
of operating conditions being considered anywhere in the country.
     The objective of a program for development of standards is to
identify the best system of emission reduction which "has been adequately
demonstrated (considering cost)."  The legislative history of section
111 and  the court decisions referred to earlier make clear that the
Administrator's judgment of what is adequately demonstrated is not
limited  to systems that are in actual routine use.  Consequently,  the
investigation may include a technical assessment of control systems which have
been adequately demonstrated but for which there is limited operational
experience.  In most cases, determination of the "degree of emission
limitation achievable" is based on results of tests of emissions from
existing sources.  This has required worldwide investigation and measurement
of emissions from control systems.  Other countries with heavily populated,
industrialized areas have sometimes developed more effective systems of
control  than those used in the United States.
     Since the best demonstrated systems of emission reduction may not
be  in widespread use, the data base upon which standards are developed
may be somewhat limited.  Test data on existing well-controlled sources
are obvious  starting points in developing emission limits for new sources.
However, since the control of existing sources generally represents
                                    2-6

-------
retrofit technology or was originally designed to meet an existing State
or local regulation, new sources may be able to meet more stringent
emission standards.  Accordingly, other information must be considered
and judgment is necessarily involved in setting proposed standards.
     A process for the development of a standard has evolved.  In general,
it follows the guidelines below.
     1.  Emissions from existing well-controlled sources are measured.
     2.  Data on emissions from such sources are assessed with consideration
         of such factors as:  (a) the representativeness of the source
         tested (feedstock, operation, size, age, etc.); (b) the age and
         maintenance of the control equipment tested (and possible
         degradation in the efficiency of control of similar new equipment
         even with good maintenance procedures); (c) the design uncertainties
         for the type of control equipment being considered; and (d) the
         degree of uncertainty that new sources will be able to achieve
         similar levels of control.
     3.  During development of the standards, information from pilot
         and prototype installations, guarantees by vendors of control
         equipment, contracted (but not yet constructed) projects,
         foreign technology, and published literature are considered,
         especially for sources where "emerging" technology appears
         significant.
     4.  Where possible, standards are developed which permit the use
         of more than one control technique or licensed process.
     5.  Where possible, standards are developed to encourage (or
         at least permit) the use of process modifications or new
                                    2-7

-------
         processes  as  a  method  of  control  rather  than  "add-on"  systems
         of air pollution  control.
     6.   Where possible, standards  are  developed  to  permit  use  of systems
         capable of controlling both  gaseous  and  particulate  matter
         emissions.
     7.   Where appropriate,  standards for  visible emissions are developed
         in conjunction  with concentration/mass emission  standards.   The
         opacity standard  is established at a level  which will  require
         proper operation  and maintenance  of  the  emission control system
         installed  to  meet the  concentration/mass standard  on a day-to-
         day basis, but  not require the installation of a control system
         more efficient  or expensive  than  that required by  the  concentration/
         mass standard.   In some cases, however,  it  is not  possible to
         develop concentration/mass standards, such  as with fugitive
         sources of emissions.   In these cases, only opacity  standards
         may be developed  to limit emissions.
2.4  CONSIDERATION  OF  COSTS
     Section 111 of the  Clean Air Act requires that  cost be considered
in developing standards  of performance. This requires an assessment of
the possible economic  effects of implementing various levels  of control
technology in new plants within a given industry. The first  step in
this analysis requires the generation of estimates of installed capital
costs and annual operating costs for various  demonstrated control systems,
each control system alternative having a different overall  control
capability.  The final step in  the analysis is to determine the economic
impact of the various  control alternatives upon a new plant in  the
industry.  The fundamental question to be  addressed  is whether  or not a
                                    2-8

-------
new plant would be constructed if a certain level of control costs would
be incurred.  Other issues that are analyzed are the effects of control
costs upon product prices and product supplies, and producer profitability.
     The economic impact upon an industry of a proposed standard is
usually addressed both in absolute terms and by comparison with the
control costs that would be incurred as a result of compliance with
typical existing State control regulations.  This incremental approach
is taken since a new plant would be required to comply with State regulations
in the absence of a Federal standard of performance.  This approach
requires a detailed analysis of the impact upon the industry resulting
from the cost differential that exists between a standard of performance
and the typical State standard.
     The costs for control of air pollutants are not the only costs con-
sidered.  Total environmental costs for control of water pollutants as
well as air pollutants are analyzed wherever possible.
     A thorough study of the profitability and price-setting mechanisms
of the industry is essential to the analysis so that an accurate estimate
of potential adverse economic impacts can be made.  It is also essential
to know the capital requirements placed on plants in the absence of
Federal standards of performance so that the additional capital require-
ments necessitated by these standards can be placed in the proper perspective.
Finally, it is necessary to recognize any constraints on capital availability
within an industry as this factor also influences the ability of new
plants to generate the capital required for installation of the additional
control equipment needed to meet the standards of performance.
     A consideration of the impact of these standards on inflation is of
major importance.  Any action which will add significantly to inflationary
pressures is considered major and requires an inflationary impact statement.

                                    2-9

-------
2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS
     Section 102(2)(c) of the National Environmental Policy Act (NEPA)
of 1969 (PL 91-190) requires Federal agencies to prepare detailed environ-
mental statements on proposals for legislation and other major Federal
actions significantly affecting the quality of the human environment.
The objective of NEPA is to build into the decision-making process of
Federal agencies a careful consideration of all environmental aspects of
proposed actions.
     As mentioned earlier, in a number of legal challenges to standards
of performance for various industries, the Federal Courts of Appeals
have held that environmental impact statements need not be prepared by
the Agency for proposed actions under section 111 of the Clean Air Act.
Essentially, the Federal Courts of Appeals have determined that "...Section 111
of the Clean Air Act, properly construed, requires the functional equivalent
of a NEPA impact statement" in the sense that the criteria "...the best
system of emission reduction," "...require(s) the Administrator to take
into account counter-productive environmental effects of a proposed
standard, as well as economic costs to the industry..."  On this basis,
therefore, the Courts "...established a narrow exemption from NEPA for
                                      1 2
EPA determinations under section  111." '
     In addition to these judicial determinations, the Energy Supply and
Environmental Coordination Act (ESECA) of 1974 (PL-93-319) specifically
exempted proposed actions under the Clean Air Act from NEPA requirements.
According to section 7(c)(l), "No action taken under the Clean Air Act
shall  be deemed  a major Federal action significantly affecting the
quality of the human environment within the meaning of the National
Environmental Policy Act of 1969."
                                   2-10

-------
     The Agency has concluded, however, that the preparation of environmental
impact statements could have beneficial effects on certain regulatory
actions.   Consequently, while not legally required to do so by section
102(c)(2) of NEPA, environmental  impact statements will  be prepared for
various regulatory actions, including standards of performance developed
under section 111 of the Clean Air Act.  This voluntary  preparation of
environmental impact statements,  however, in no way legally subjects the
Agency to NEPA requirements.
     To implement this policy, therefore, a separate section is included
in this document which is devoted solely to an analysis  of the potential
environmental impacts associated  with the proposed standards.   Both
adverse and beneficial impacts in such areas as air and  water pollution,
increased solid waste disposal, and increased energy consumption are
identified and discussed.  Appendix B of this document outlines those
sections or chapters which examine these potential environmental impacts
in detail.
2.6  IMPACT ON EXISTING STANDARDS
     Standards of performance may affect an existing source in either of
two ways.  Section 111 of the Act defines a new source as "any stationary
source, the construction or modification of which is commenced after the
regulations are proposed."  Consequently, if an existing source is
modified after proposal of the standards, with a subsequent increase in
air pollution, it is subject to standards of performance.  [Amendments
to the general provisions of Subpart A of 40 CFR Part 60 to clarify the
meaning of the term modification  were promulgated in the Federal Register
on December 16, 1975 (40 CFR 58416).]
                                    2-11

-------
     Secondly, promulgation of a standard of performance requires States
to establish standards of performance for existing sources in the same
industry under section lll(d) of the Act if the standard for new sources
limits emissions of a pollutant for which air quality criteria have not
been issued under section 108 or which has not been listed as a hazardous
pollutant under section 112.   If a State does not act, EPA must establish
such standards.  [General provisions outlining procedures for control of
existing sources under section lll(d) were promulgated on November 17,
1975 as Subpart B of 40 CFR Part 60 (40 CFR 53340).]
2.7  REVISION OF STANDARDS OF PERFORMANCE
     Congress was aware that the level of air pollution control achievable
by any industry may improve with technological advances.  Accordingly,
section 111 of the Act provides that the Administrator may revise such
standards from time to time.   Although standards proposed and promulgated
by EPA under section 111 are designed to require installation of the
"...best system of emission reduction...(taking into account the cost)
..."the standards will be reviewed periodically.  Revisions will be proposed
and promulgated as necessary to assure that the standards continue to
reflect the best systems that become available in the future. Such
revisions will not be retroactive but will apply to stationary sources
constructed or modified after proposal of the revised standards.
2.8  REFERENCES
1.  Portland Cement Association vs. Ruckelshaus, 486 F. 2nd 375
(D.C. Cir. 1973).
 2. Essex Chemical Corp. vs. Ruckelshaus, 486 F. 2nd 427
(D.C. Cir. 1973).
                                  2-12

-------
            3.  THE STATIONARY GAS TURBINE INDUSTRY AND PROCESS

3.1  GENERAL
3.1.1  Introduction
     A gas turbine is a rotary engine, of which a common example is the
aircraft jet engine.  The hot combustion gases from a stationary gas turbine
drive a power-output shaft rather than providing thrust as on the aircraft
engine.  Figure 3-1 presents a cut-away view of a gas turbine.  Turbines
range in size from less than 40 horsepower (0.03 megawatts (MW)) to over
one hundred thousand horsepower (75 MW).  Manufacturers continue to increase
the horsepower of individual turbines and frequently turbines are "ganged"
or installed in groups, so that the combined power output from one location
may exceed 1,648,000 horsepower (1230 MW).
     The use of stationary gas turbines has increased tremendously since the
mid-19601s.  Several characteristics of gas turbines which contributed to
this growth are:
     *  can be instrumented for remote operation and, therefore, can be
Installed almost anywhere;     /
     *  Time from placement, of an order to on-line operation of a turbine
Is relatively short;
     *  quick and easy installation;
     *  high horsepower to size ratio;
     *  short start-up time;
     *  relatively vibration free operation;
                                      3-1

-------
_Q
 s_
 to
 re
 Ol
+J
 re
 re
 o
 re

4-
 o
 re

 re
o
 i
oo
 O)

-------
     *  good reliability;
     *  capability of operation with a variety of fuels in any environment;
     *  operate without cooling water;
     *  have lower capital costs than diesel engines;
     *  have a very low physical profile (low buildings, short stacks,
little visible emissions, and quiet operation when properly muffled).
                                   2  I/
3.1.2  Gas Turbine Industry Profile   -
     This section discusses the basic structure of the gas turbine industry.
It contains analyses of the various market sectors and growth projections
showing the type and number of units likely to be affected by the new
source performance standards.
     The analysis is divided into five sections:
     (1)  description of turbine manufacturers;
     (2)  market description;
     (3)  historical market trends;
     (4)  market projections by application; and
     (5)  effect on the balance of trade position.
The analysis uses 1974 data for the thirteen manufacturers and packagers of
turbines.  All information presented, unless otherwise indicated, is in terms
of orders placed rather than equipment delivered or installed.
3.1.2.1  The Manufacturers and Major Sellers of Gas Turbines - The United States
stationary gas turbine industry consists of thirteen firms.  Of those firms,
only eight actually produce their own gas turbines.  Table 3-1 lists the eight
manufacturers and five packagers of stationary gas turbines.  Cooper-Bessemer,
-   Unless otherwise noted, Reference 2 is the source of all market information
in Section  3.1.2 and its subsections.  Specific references for statements in
Section 3.1.2 are contained in Reference 2.
                                      3-3

-------
















t/0
LiJ
z:
CQ
Ct

P
(/)
^£
CJ3

01

C
C

CT
Ul





a.

o
§



o.
:c
o
eo
VO




C g
Cf. C

O U.
o
LO +J
c
01 O)
c t.
3 re
•P Q.
i.
o **-
LL. O










1
tZ
4J
c
01
s-
o.






I






r—
CM









r_
re
c

to
c
o
•!-> U
S- *— »
i- in
c
•)-> n>
+j a.
6* o
I- O
3




AiResearch Manufacturing Co.,
Phoenix, Arizona


3:
Q_
3: c.
o
CO ^*
r^ to
to r,
to

i
O-
o- 3:
3:
o
O LO
0 O
LO CO








co r^
CO O
CM «1-










t
U
c

LO
CU
iL
>> m
C 3
re -o
*
o o.
1 1




Avco Lycoming Division,
Stratford, Connecticut
Cooper-Bessemer Company—'
Mt Vernon, Ohio



-Jf


s
00
1

Q_
0
LO

co








CJ

LO















•u
Ol
•r"
s-
3t
in
in
»^-
4->
t-
3
O




Curti ss-Wri ght Corporati orn-'
Woodridge, New Jersey

a.

r%
3: o
0
LO LO
CO CM
tt

1
D-
a. 3:
o 8
tO LO
CM r—








CM to

r—








C
O
+J O
re c
o '~l
Q. •
8 1
in •(->
t- in
1 1
r— fc.
re 01
I- in
0) in
C 01
0) »•
c5 o



M
C
O
Detroit Diesel Allison Divisi
Indianapolis, Indiana
Dresser-Clark Division-''
Houston, Texas

2
X

LO
.
2 S
CO
Q
* 1

1 O,
3:
2 8
LO O1
"~ "*








co
i
i










c
re


u
t.
LU
re
n\
\ Ql




Energy Transformation Corp.,
Bayertown, Pennsylvania
General Electric Company
Schenectady, New York

o.
3:




CM



0
to
CM









CM
M-













C
re
o-

•D
1
i
o
c
w
c




Ingersoll-Rand Company-
Phil lipsburg, New Jersey




D-

0 *
OO O
00 CO
co «•

1 1
3: i
o to
•* to








 en
E £
5j=
+>
t.
i *
E *
|Q HJ
E |

c 5
I-" 




Solar Division
San Diego, California
Turbodyne Corpora tion-
St Cloud, Minnesota





o
LO
CO
1

O-
x
§
01

"








to
•a-









c
o

re
s.
0
Q.
t-
8
if.
re
5

•i*
I




Turbo Power and Marine
Systems, Inc.
Farmington, Connecticut





~m
I
JO



i
__
•-









^^





c
0

s
o
Q_
r
8

u

u
LU
01
V)
en
c
. fc
in



1
Mestlnghouse Electric Corpora
Lester, Pennsylvania








4-*


._
^L
s
>*-
0
cu
01

CL
Cu
c
01 O
V)
re m


re o)
-U r—
re o
•o X
0
X)
o 01
c c
t— t *l"
1
S 8
y
•^ re ,

01 1-
•^3
D> C 01
•r- T- 3
01 f- in
t S *
JO ¥>
X to oj
at- t
3O1 3

fu o>
01 O 41
£.•— r-
t-»- <0 Q.
O) Q *J i-
C « +-1
•»- Q O r—
•r- ^ O 3
P" C c:
Of 1— 0) i- in
c in 43 "C re
0> ib 3 => 01
> <3ol- C
••- o •*•>
j. s in
13 i. c O "oi
01 t- 3 N
f- T3 i. - in
-e afifc *
5 «<«£ i
^ .c re B
o •**•
3
S- CT
0* in tt)
C7> Ol *
re o
^ l_ in
O 3 i-
^i Sri
3-4

-------
Curtiss-Wright, Dresser Industries, Turbodyne, and Ingersoll-Rand package
and sell turbine units that they purchase from some other company.  The
English firm, Rolls-Royce, is the major supplier of gas turbines to four
of the firms while Turbodyne packages Brown Boviere turbines.
     The industry is dominated by giants.  As shown in Table 3-1, seven
of the eight manufacturers are owned by companies in the Fortune 500 rankings;
and five of the eight are among the 50 largest manufacturing firms in the
United States.  The table also provides information on the size range of the
firms' currently available equipment.  The wide range of sizes indicates
that some of the firms compete in different markets.
     Table 3-2 ranks the turbine manufacturers and packagers by 1974 horse-
power output.  Because turbines vary widely in size, horsepower provides
a better surrogate for dollar volume sales than number of units sold.  The
industry is very concentrated with the three largest companies holding over
80 percent of the market.  The corporations manufacturing stationary gas
turbines are all highly diversified, and six of these also produce recipro-
cating  (i.e., diesel) engines, the gas turbine's major competitor.  It is
interesting to note that the top three gas turbine manufacturers  (General
Electric, Turbo Power and Marine, and Westinghouse) do not produce recipro-
cating engines as part of their overall company organization.  Any regulatory
change, therefore, that seems to favor reciprocating engines relative to
turbines might be most strongly opposed by these companies.
     The market for gas turbines is worldwide.  In 1974, approximately half
of all unit sales and 30 percent of all horsepower manufactured was sold
outside the United States.  While the regulations under consideration apply
only to turbines sold in the United States, changes in the manufacturing
process could also increase the price of turbine exports.
                                       3-5

-------












t/1
ce:
LJL)
Q
rv^
O

1
O
H-

CQ

Q
UJ

Z

S
oo

i — i
u~
UJ
I-H
CQ

— )
t-
OO
3

>_
o:
«a;
o
»— i
i —

j—
oo


,
CM
1
CO

cu
ro
h-




















t/1
O)
E E
O -r-
•i- CT>
•M 10 E
rO V. LU
i- CU
O J- Ol
Q. 3 E
S- 4-> T-
o o ••->
O ro ro
»4- 0
+•> 3 O
E E t.
CU ro Q.
s- s: •-
(O O
O- CU
ct:



<3- to
r*-> _|j
- ^—













^" C3 *ct" f^ CT*
O «d- CO O
r— LO













CO tO CT> r— CO
•«*• co cr> LO *i-
CO CM i—

















r^* co o *^" CTI
CO «i- VO O O
«d" O CM CO CO
CM CM to in co
VO LO CO







o

^_3
o cu
CU t""^l

iii O Co
J= E
i— en >>^~,
rO S E -O CO I
S- •!- O i-
CU 08 +-> J3 ro
E c/) t- ^~
cu a. cu 3 o
C3 1— 3 r— 00









V) O O to
cu z z cu
>— ^M>













•a: *t -a- r-
Z CO •—














< CO r— O1



















CO ^^ tO
 CU
O cn si E
to c cn cu
CU •!- -r- V)
•r- E J- t/>
a o 2 cu
O 1 CQ
4») 2>) (/) 1
O 1 -r- CU
S- O +-> O-
4-> O S- O
CU > 3 O
o > r— «*
CM co cn m
in co «d" i—
"~ ^.
f**^
r™"

E
0
•r—
4J •»->
4J ro
-^, CU E
•M ! •».. J- C
-OCM| J- o
E ^ (O («- OO
us s- CD to _j
DC ro ^>t. E ^C
1 r— -E ra h—
r- O 0 S- O
r- 1 i. t~ 1—
O S- ro
CO CU CU >,
S- CO OO CD
cu to a> t-
cn cu a: cu
E i- T- E
(-H O 
E
cu
ex
••"
to
M-
0
CU

rO
•o


j2;
cn
i—
**-^

. r^
to cu
CU T3
S- 2
3 4-> E
CT E
•i- a> a.
4— E HT*
O-
r— -r- O
r— 3 in
ro cr co
cu
S- S-
<4- CU <*-
•O i- -O
0) J- 01
«/) "O to
3 3
CU
*o c *o
•M »i- -4->
fO /^ > cu >>
E cn E
ro rd ro
Q. ^ a.
E U E
O Q. O

•^^ "^^ "^*
i— I CM) co|





















cu
to
.a


rO
"O

"
0
E

cu

i-
cu
CO
CD
E
•r-
•»->
cu
^
tw
ro
yr
f
L.
CU
^H>
3
Q.
E
0
0

cu
E
•i—
_a
3
to
rO
CO

E
O
i.
t|_

•a
cu
a.
o
T3



• •

-------
3.1.2.2  The Market for Turbines by Equipment Size Range - Gas turbines have
an extremely broad size range that starts under 100 HP and approaches 100
megawatts (134,000 HP).  Generally, units over 20,000 HP are used for electric
power generation and units under 20,000 HP are used for a variety of purposes
including gas compression, stand-by power and miscellaneous Industrial uses.
     Table 3-3 shows the 1974 stationary gas turbine market divided by equip-
ment size.  Over 70 percent of the units sold In the United States are less than
5,000 HP while over 85 percent of the  HP sold is in unit sizes of 50 MW
or greater.  The mid-market area from 5,000 HP to 50 MW contains less than
10 percent of the horsepower and units sold in the United States.  This
market area, however, is more significant outside the United States, accounting
for over 20 percent of horsepower sales worldwide.
     Table 3-4 presents the firms competing 1n each size range showing 1974
domestic orders by units and horsepowe'r.  The table Indicates that there
are two competitive groupings.  The smaller firms compete in the market for
units less than 5,000 HP while the four largest firms dominate the remainder
of the domestic market.  Only one packager, Curtiss-Wrlght, competes in the
large horsepower end of the market.
     The size categories can be divided into four groups as shown below:
                        Turbine Sjzes and App]1cations
     Size Category       Equipment Size       Number of Finns Competing
                                              Mfg.                  Pkg.
     Small Capacity      5,000 HP              2"                    1
     Medium Capacity     5,000-20,000 HP       2                     3
     Large Capacity     20,000-134,000 HP      3                     2
     Multiple and Com--/   134,000 HP          2                     0
       bined Cycle
These groups are unique in terms of competition or types of applications.
-  See Section 3.2.1.3 for definition of a combined cycle turbine.
                                      3-7

-------
oo

O
oo
          >-
          t—
          1—I
          o
     UJ
     o
     OL

     °   5
     UJ
     :r   a
     i—i   Zt
     ca   «=c
     ex:
     rD   oo
     h-   H-

     oo   2:
CO


CO
       M
       O£
       o
                                       00
              a;

              f
              «1
                    «|  yj


                   OO|  -r-
                                                 o
                                                 CM
                     (/l
                     s~
                     ai
                     -o
                     S-
                     o
   S
   UJ
   QC

   CO
                                         CO
                                         **•
                                         00
                                         S
                                         0
                                                                                   tr>

                                                                                   oo
                                                                   CTl


                                                                   00
                                                                    LO


                                                                    CM
                                              S
                                 o       r-
                                               CM
                                               r-
                                               °°
                                                  10
                                                   00
                                                   00
                           01
                                  o
                                  0
                                  0
                                                2
                                                g
                                             I



                                            O
°
o
o
                                                            00
                                                             Q.
                                                                                    ro

                                                                                    10
                                                                         OO
                                                                         ^1-
                                                                         O
                 CM
                 CO
0
0
o
o
CSJ
                                                                           O
                                                                           O
                                                                           O
                                                                                                      00
                                                                                                      cr>
                                                                                                      ro
                                                                                                       "*•

                                                                                                       "
                                                                                      10

                                                                                      U3
                                                                                      en
                                                                                   IO
                                                                                   CM
                                                                                   in
                           IT)  I

                           CM
                              IT)

                             I i—


                           LT> O
                           r- CM
                                                                                             oo
                                         CO
                                         cn
                                                                                              CM
                                                                                              CM
3 LO
So
   r~
O IO
ir>
    i
                      •— CM
                       . If)
                      ID CO
                      CM CO
                                                              CM
                                                              CO
                                                                                                                   cn
                                                                                                                           O
                                                     O
                                                     o
                                                                                                                                o

                                                                                                                                o
                                                                                                                                o
                                                                        o
                                                                        o

                                                                                                                     O
                                                                                                              O
                                                                                                              oo
                                                                                                          10
                                                                                                               CM


                                                                                                               52
                                                                                                        O 1 --
                                                                                                        tO 10
                                                                                                                               S
                                                                                                                      vo
                                                                                                                               00
                         o «d-
                         O PO
                                                                                                                                           a>
                                                                                                                                           
                                                                                                                                           re
                                                                                                                                            
-------
ri

ID
      OO
      LLJ
      CO
      Qi
      oo
      <
      0
          O


          O
oo
or
      Q
      D£
      O
           C3
      u_   oo


      Ll_   >-

      O   CO


Q.
Tt CO 3:
ic o
o • 3; co
o «*• +->
1 	 CO •!-
i — C
^
ZI Q-
CM 3Z
O 1 O
O CM i— D-
,— un CM 3; co
1 O CO -I-)
r"~. ciS c
0 =>
UO
•3: o-
Z 3:
i i —
o r^ un a.
in CM O 3Z
i in i"-. co
. — ro co +•>
. co •<-
m c
CM ZD
O-
3 31
s: i co
in CM O-
tn r— un ar co
CM .— OO 4->
1 O CO •!-
un CM c
i— rD
0
O 0,
o in
o
CM d.
1 -'
i — CO
0 +->
o -r-
o c
•— •=>
o
O Q.
O X
0
i — Q.
1 X
i — CO
O •!->
O -r-
in c
=>
O D.
0 3;
0 Q-
m a;
1 CO
i — +->
0 •<-
0 C
CO 3

O D.
o 3:
O Q.
co ac
i t/
i — 4->
O •!-
o c
D.
O 31
o
O 0.
^ — J—
1
i — CO
4^

C


^
U-






o
o^
o
in
CO

0
«*
«*
co'
o
f—

•cl-



lO
CO
cri


CO

! 	
.
r».
O


CO




CO
03



^—


















o

1— c
10 •(->
i. U
01 CU
C i —
CU LU
o
0
CM
Cn
oo
CO

r^
*a-
co
o
en


CM
i —
, 	
co"
p—

co









o
.
o
10


CO






























z:
09
O.
I—
UO
r—
ro
r»-

co
CM
CD
r-»
CO


00

o
in
•*
^-*

CO













































1
CT)
C O)
•i— CO
•(-> 3
co O
CD -C





**•
"3-
00
«*


CO




















































-x.
t—
O 0)
ja c
i- >>
3 T3
1—













































CO
•
CM
t-^
r™
m
*!•

m
•
CO
CM
CM

•*f
oo
CO
•
r~.
^r

in
CO
1



^
10
o
oo













































«3-
CO

^

«*
•
f-.

CO
CM









O)
C.
•r—
U
>>













































<
^.
















C
•4-J r- O
•i- 0) 
10 •»->
cu cu
tO S-
 i-
o: ie
•i- C3
•a:













r—
CM
CO
r—

co



























0
CM
in

CO













to
S-
dl
s_ en
CO O IO
s_ -r->^:
QJ IO O
jc s «
-M 	 Q.
O
CO

co
CM
CO
•rt-

O
CM
O
•
r^
CO
r*^
LO
CO
CO
vo
co'
CTl
«d"

CO
r^


CO
CO
0>
r—


CO

r—
•
r^^
CO


^~
f~~


CO
CO




r—
o
•
CO
CM
CM
cn
in

en
•
r-~
01
CM

0
CM
O
r^
T^

OO
r--.
r—



OO
<
o
1—
                                                                  3-9

-------
     3.1.2.2.1  Small  capacity gas  turbines  - Those gas turbines under
5,000 HP are classified as  small  capacity  gas turbines.   In this size
category, the r'jority of units  are used for and  divided  fairly evenly
between gas compression and stand-by electrical generation.  Over  70 percent
of all  units sold in the United  States  in  1974 were under 5,000 HP.  However,
these units represent less  than  5 percent  of the  total HP capacity of all
units sold in the United States  in that year.
     Under 1,000 HP, the AiResearch Manufacturing Company, a division of
Garrett Corporation, dominates  the market.   In the last two years, the
company has increased its domestic sales by  over  300  percent by orienting  its
sales to OEM-  contracts.
     From 1,000 to 5,000 HP, Solar Division  of International Harvester  has
increasingly dominated the market for at least the past five years.  The
overwhelming majority of its units are either gas compression  or stand-by
electrical generation sets.  Solar's main  competition comes from Detroit
Diesel  Allison (a division of General Motors).  An industry magazine reported
that between 1970 and 1974, GM's gas turbine sales jumped 45 percent.
Solar's sales during that period increased at approximately the same rate.
     Avco Lycoming is the only other manufacturer actively competing for
the small capacity gas turbine market.   Its  1974  production was approximately
20 percent of Solar's with sales concentrated almost  completely in units
under 3,000 HP.  Solar, Detroit Diesel  Allison, and Lycoming did not sell
V  An Original Equipment Manufacturer (OEM)  is one who buys  an  engine from
an engine manufacturer and incorporates it into a  product of  which the engine
is only a component (e.g., a portable compressor which is usually assembled
by an OEM using his own housing and compressor but a purchased engine).
                                    3-10

-------
any units over 5,000 HP 1n 1974.   However, Solar 1s building a 10,000 HP
turbine which should be on the market soon.
     Of the packagers, only Ingersoll-Rand sold any units domestically
in this size range during 1974.  The company sold twenty-two 4,000 HP
Rolls-Royce gas turbines in 1974, thirteen of them domestically.    Cooper-
Bessemer advertises a unit in this size range (also Rolls-Royce)  but there
is no record of the company making any sales under 12,500 HP since 1972.
The recent drop in domestic gas compression sales probably accounts for
the limited Cooper-Bessemer sales.
     3.1.2.2.2  Medjum capacity gas turbines - Mid-range gas turbines, between
5,000 and 20,000 HP, are employed in a variety of applications with the
majority utilized in stand-by generation and pipeline compression and pumping.
However, this is the size range where gas turbines are diversified in their
applications.  These include mechanical drive, shaft power, and a variety
of other industrial uses.  This category accounted for only a small part of
the total domestic sales in 1974.  As electric generating units have
increased in size over the last five years, the relative demand for this
size gas turbine has dropped.
     General Electric, TP&M, and Westlnghouse are the only U. S.  manufacturers
of gas turbines that were active in this range 1n the last five years.  In
1974, only GE and TP&M sold any units domestically.  Three packagers,
Ingersoll-Rand, Dresser-Clark, and Cooper-Bessemer exported 21 units in 1974.
This is more units than the major manufacturers combined sold in  1974 and
these packagers have been actively competing over the past five years.
     This size range is one area where foreign manufacturers have made inroads
into the American market since 1970.  Private industry purchased  five units
                                       3-11

-------
totalling 69,300 HP 1n 1974 for electric power generation,  and  In the last
five years, several foreign firms have sold American oil  companies gas
turbines to be used 1n pipeline compression.
     3.1.2.2.3  Large capacity gas turbines -  Gas turbines  over 20,000 HP
or approximately 15 megawatts (MW) are almost  exclusively used  in electric
gen jrator drive applications.  The largest single unit gas  turbines in this
category reach sizes that approach 100 MW or approximately  134,000 HP.
Between 15 MW and 100 MW, active market competition occurs  between five
American firms.  These firms are General Electric, Turbo  Power  and Marine,
and Westinghouse; in addition, Curtiss-Wright  Corporation and Turbodyne
Corporation package and sell turbines under their own names.
     Because of the considerable expense and complexity of  the  equipment
involved, these five companies produce gas turbines only  on a custom order
basis.  The lead time for delivery in 1974 was from 6 to  9  months, but as of
May 1975, it had climbed to 24 months.  This 1s much longer than that required
for the smaller, mass-produced models discussed earlier.
     The custom building of this large equipment allows for variations in
rated generating capacity for each model sold  by the five firms.  The
active competition in this market has produced increased  efforts to improve
fuel efficiency levels.  In addition, newer techniques have been developed,
such as combined cycle technology.
    3.1.2.2.4   Multiple units and combined cycle - Since the late 1960's,
growth in total HP installed has been accompanied by a substantial drop
in the number of units produced.  This trend is partly due  to the industry's
increasing production of multiple tandem and combined cycle installations.
These installations have the capacity to produce well over  100  MW with one
                                      3-12

-------
                                             3
application having a total rating of 1230 MW.   Table 3-1 indicated that
five companies market individual equipment configurations that have a
generating capacity of over 350 MW.  Multiple unit installations with
these large capacities are increasingly being installed for mid-range
(3,000 to 5,000 hours) and baseload capabilities though they still represent
only a small portion of the total market.
     In 1974, TP&M, GE, and Westinghouse were the only firms to market
installations with generating capacities greater than 100 MW.  TP&M
was the solid leader selling 17 installations with a total combined capacity
of 3.9 million horsepower.  In addition to these three firms, over the past
five years, Pratt & Whitney has also recorded sales in the 100+ MW category.
3.1.2.3  Historical Market Trends - The domestic production and utilization
of gas turbines blossomed in the 1960's after a period of experimentation and
development in the 1950's.  This growth for the last ten years is indicated
by the graphs in Figures 3-2 and 3-3 which show yearly domestic orders for
gas turbines in terms of units and total horsepower sold.  Earlier data
indicates that production expanded sharply in the early 1960's, but after
1970, as shown in Figure 3-2, the number of gas turbine units produced
declined drastically.  Total HP sold, however, stayed relatively constant,
reflecting the use of gas turbines for utility electrical power generation.
In general, these applications require larger units and this factor,
combined with a steady decline in gas transmission sales in the United States,
accentuated the inverse relationship between units sold and HP capacity.
     The graphs indicate the dominant role played by utilities in the
existing market.  Over 90 percent of the horsepower sold in the United States
goes to utilities and that margin  continues to widen.  Turbines for oil and
                                    3-13

-------
                                          Figure 3-2.

                        YEARLY  DOMESTIC ORDERS  FOR  STATIONARY GAS TURBINES
                          BY  APPLICATION,  GIVEN IN  UNITS,  BY YEAR
                  Legend;
UNITS
SOLD
IN
HUNDREDS
           10  r
                   m
Utilities

Oil and Gas

Private Flectric
Power Generation

Stand-by

All Others
                                             68    69    70    71
                                                    YEAR


         SOURCE:  Adapted from Gas Turbine Computer Marketing Service, Inc., data base.

                                                3-14

-------
                        Figure 3-3
 YEARLY DOMESTIC ORDERS FOR STATIONARY GAS TURBINES, BY
        APPLICATION, GIVEN IN HORSEPOWER, BY YEAR
12 r
11
10 .
HORSE-
POWER
IN 9 •
MILLIONS
•
7
6
5 .
4
3 •
2
1

]] Utilities

	 ui i ana \i
:':'•:•: Private E
Power Gen
x^ Stand-by



as
lectflc
eratlon
^ All Others
. '4
. \
1 1; .1 !! '.

F8S'''{^^\
^v^xf XvX^
I 	 _- 1^~ 	 ~ —

•
i
11
! '•
i
' i
l .
f:'4:^.
1
1
i;
1 !
j
liiii Hi-
| ! ; j , ; i
1 1 . | ' • ! t
: » i I i i.i-
'|-:j 1
•"i1! ! Ji
, J , • 1 --L —I

H 1§

T""
i
t i
i i
i i ,
j i
1 i l
; J '
' 1 ' i
i (
i •
1 1 i i
1 i » i
' ! ; i i
1 • . • ' i
i * ' |
"' -i
1 !
1 1 1 i
, i i : '
' ' i
i ; h;
• i ' •;
1 i i -i
! j 1 j 1 ' 1
; ; i '• i i ; i i
i • : i • i
. ' i '
. : i . . ;
' ; ' . i i
1 i i « •
. 1 • i >
1 , I • • '

	 • • •
'^'-'-'-•^-.--

\
i . i
. i
> . i •
l ' '•
I I
i • !•
4 ,
: ; i
. i i
i ' '
i •
! i.
1 ! • '
,' :
1 ;
• 1 i ! • I


1
1
'• 1
1
(1
IP
i i
1 • »
; .
• j i : '
. . i ' '
: i , ;
; i • : ;
: ! '
i
'••ii'
• t; ;
'••III'
1 !
ill .1
1 i ' i '
' • i ! '
i • , '
! J
t '
j
i
i
*
i 1
' l
1 *
1 ' l
1
; i
: jj
! i i
- 1
! j
; i l
•i
:: :
• * *
i
i •
i - * .
: l ' j
' ' i
' 1
• ' i
1 ! '
: li
ii ':'
• 1
i :
t *
l • ';
t ;:
' : j i;
M;
•'.I ii
— c— '-
     65
66    67    68
69    70    71
72
73    74
                            YEAR
Arapteci from Gas Turbine Computer Marketing Service, Inc, data La-
                                 3-15

-------
gas, stand-by electricity generation,  and utilities  comprised almost 90
percent of the unic sales of turbines  1n  1974.   Unit sales  for all  three
applications have declined since 1970  reflecting both larger turbine capacity
and declining demand.
     Figure 3-4 presents the cumulative growth  of domestic  gas turbine
pro'uction in terms of the number of units and  total horsepower ordered.
Again, the graphs indicate that while  total installed HP continues to
increase rapidly, unit orders have beer, decreasing since 1970.  The
figure also emphasizes the dominant position which utility  applications have
achieved in the market over the past ten years.  Because of this dominance,
recent innovations in turbine design have focused on the utility market segment.
     One market event deserving special consideration 1s the oil embargo of
1973.  Turbines are less fuel efficient than diesel  engines and are sensitive
to the price of fuels.  When the price of oil Increased dramatically as a
result of the oil embargo, the market  for turbines declined.  Figure 3-3
indicates 1974 turbine orders for 12.4 million  HP but many  of these orders
have since been cancelled.  Robert Farmer, Director of Computer Marketing
Services for Gas Turbines, C.M.S., Inc., estimates that preliminary data on
cancellations for 1974 could drop the  existing  estimates by 50 percent.
While certain markets remain open to gas turbines, those where fuel costs
remain high relative to capital costs  will likely decline first.  The
capital costs of turbines are one-third to two-thirds less  than reciprocating
engines; however, they may consume as  much as thirty percent more fuel per
horsepower output.
     In 1972, gas turbine market analysts predicted a continuing growth
in the importance of gas turbines, especially in electric utility applications.
                                     3-16

-------
CO

0)

3
     3
 O
 t—«
 " '  UJ
 oi  >
 O  •—'
 u.  CD  a:
o£  s.-  o
S  s  £3
§  fe  S
    o  o
(_) ' I—I  I
H-*  —1
h-  o-  a
            •o

            CL)
            CT.i
            (1)
                                                                                         •1"
                                                                                         ii
                                                                                          .!\«l
                  DLf—
                  (U-r;
S
    CO
UJ  tO
>•  UJ
1:1
             o
             §
                 O) I/)
                                 O
                                 o
                                 o
                                  *
                                 00
                                            o
                                             A

                                            VO
                                                                                           .li^
fc»— «
0
o



__ .
§
c
•r™
(/)
fc. c
0) O
'-
S 5



... 	 	 	 ~"
> o
F <*



\
                                                                           \
                                                                             \
                                                                                       l
                                                                                      v
                                                                                            CM
                                                                                         IM r-
•111
\
\'
\l
\\
11
l"
ill
it
V 1 '
\ 1
MI
• r>.
o
r«*
en
vo

00
VO

r-.
vo


vo
vo

01
S
>-








                                                                                             LT
                                                                                             vc
                                                                                    '  M V

                                                                                    \\\\
                                                                                                      a>
                                                                                                      10
                                                                                                      JD
                                                                                                      (O
                                                                                                       (O
                                                                                                      "O
                                                                                                           o
                                                                                                       0)
                                                                                                       o
                                                                                                       o>
4->,«
(O Z
                                                                                             CO
                                                                                             P-.
                                                                                             CM
                                                                                             CD   UJ
                                                                                             VO   >-
                                                                                             oo
                                                                                             vo
                                                                                                 vo
                                                                                             vo
                                                                                             vo
                                                                                             in
                                                                                             vo
                                                                                                     «' i
                                                                                                     e «<

                                                                                                     10 -i

                                                                                                     0) J
                                                                                                     XJ

                                                                                                     o •
                                                                                                        •
                                                                                                     *±
                                                                                                     r-
                                                                                                     cr.
    o
     •t
    C«J
                                                3-17

-------
Those experts saw the potential  for a $2+ billion market by 1980,  nearly
triple the $800 ,,.illion 1n sales of 1972.
     The national fuel crisis changed this prognosis as utilities  took
a wait-and-see posture.  Orders  slowed to a fraction of their previous
levels.  Because of the current  24-month lead time, the impact of  the
Initial drop in orders has been  delayed.  With the 2-year order lag, 1976
orders will not be shipped until 1978 or 1979; therefore, even 1f  the market
turns around, manufacturers must be prepared to carry losses an additional
two years before they can hope to again be generating profits from gas
turbine sales.
     This lag in business will have a major impact on the small component
suppliers for gas turbine parts.  They do not have the financial base of
a GE or Westinghouse, and two years of slack demand could cause many to go
out of business.  This, 1n turn, would reduce the number of component
suppliers available when orders  do pick up.  Lack of suppliers would lengthen
the current Installation lag time and potentially raise prices.  The net
effect of all these factors would be to reduce the gas turbine's competitive
edge in the markets to which they currently appear best suited.
3.1.2.4  Market Projections by Major Application -
     3.1.2.4.1  Overview - The major applications for gas turbine equipment
can be divided into five broad categories:  electric utilities; oil and gas
industry; stand-by electric generation; private electric power generation;
ind other industrial applications.  These categories are analyzed in greater
retail in this section.
     3.1.2.4.2  Electric utilities - The most significant use of gas turbines
today, in terms of total HP installed, is for electric power generation by
                                      3-18

-------
      electric  utilities.   The  relative  position  of  utility  applications  to other
      gas  turbine  applications  has  Increased more than  10-fold  over  the past  10
      years,  an Indication  that gas turbines have become  an  essential part of the
      nation's  power base.
           Gas  turbines  are considered most popular  as  peak-load  equipment.   A
      random  sample of gas  turbine  applications taken from a 1972 Federal Power
      Commission source  Indicates  the following relationship between hours run
      connected to load  and percentage of units Installed.   Most  units run less
      than 2,000 hours per  year.
?/>   30 r-
    20
UJ
Qi
UJ
D-
    10
                   2000            4000            6000
                             HOURS ANNUALLY CONNECTED TO LOAD
                                                                  8000
           Shipments of gas turbines to domestic electric utilities are presented
      in Table 3-5.   Shipments peaked in 1971  after a dramatic 13-year climb.   The
      major decline in shipments, however,  began in 1974 after 1973 fuel  price
      increases discouraged turbine buyers.
           Estimated 1975 shipments are 62  percent below 1974 levels and 72 percent
      below 1973 levels.   In 1973, 20 percent  of all  orders for new generating
      machines were for gas turbines, but by 1975, this figure had fallen to only
      seven percent.
           The cost of turbines per kolowatt correlates with volume because the
      fixed costs of turbine manufacturing  (e.g., set-up costs) are so high.   The
                                           3-19

-------
                     CU
                     c  cu
                    •r-  CT
                    .a  nj
                     S- 4->
                     3  C
                    _CU

                     10  S-
                     13  O>
                    CD CL.
                                      -•»   O  CM  r—
                                                                                                                                  in
                        to
                    <4- 4J
                     O  C
                        cu
                                 COVOCMCT>CT>COCMVOi—  CMVOCOCMLnCOOO
                                 *tCOi—   r«-OCM«J-CMVOini—   O«3-CMCMOO
                                 ococnco^cnr^cMOr—  lococoin^r—  vo

                                 CMi—   r—COr—   CMCMincOOOcyii—  CMOr>»COr—
                                 i—  i—   r-       i—   i—   r-  r—   COCMCMCO«efrinCOOOCO
            CO
      CO    LU
If)    I—I
      oc
OO    CO

 CO    LU
m    ca    (~>
|~    Oi    LU
      CO    CJ
             CO
             LU
        ;   8
                     J-  to
                     CU  4->
                    JC  C
                        O.
                       -r—
                       J=
                       CO
                        CO
                     CU  C
                     3  O

                      O> (*•••. CO 00 O O i— 00 P"**. 00 ^~ CO O
^- 0*( r«» ^^ 00 Cft CM CM f^ CO C^ CM CM CO ^~ CO CO
o r^. co <£> r- Lnco^oi^.iocooocoovom
CMi~i— cor-cMCM^-o^frtniotoooor^a^
f"~ • ^™ (•*•* p™* r^ f^ r~~ CO CM CVJ CM OO ^^" CO CM CM










CO 00 VO f™* ^*™ CM t-^ ^^ Lrt L^ CM ^^ ^^" ^™~ CO CO
1 ^•COr*-.O^COOCOCMCMO>^O r— CM CM ^" LO
f_« f**m f^ pO ^^ tp PO *!lj" |*™» CO p"» 00 ^^« L^
r*"* CM CM *«^ L^^ LjO t^ CO ^~™











1 ^OOO^*r^^-CMCOr— lOr^CMCOi 	 LOOOLO
i o^ oo oo co ^""» f^* ^^ ^» vo *jO ^** vo r*** UD ^D r*^










CM| CM| CM! CM!
^t-Oi— CMCMi— CTiCTiCMCMOOr— r— i— OO
i^cMCMco^'incM'ni^r«»oocM^J'inr—
CMCocomrx.f»r^.incM







aiOi — cMco^j-invor^oocDOi — cMco^m
(O •
— t/)
4->
ti ^"
*T^ ^.
o , ex
o cu
Ol i—
cu o
•4-^ ^*>
(O . O
o o
i- ID
•a cu
• • >> c
to ^: -I-
C TJ E
•i- C O
4-> 
c t-
0*0
cu a.
cu TJ cu
CO >, C

CX-i-
D--Q CO
•r*" ^^ t5
JC O C7>
to o
CU
(/) <+- JT
- O 4J
s: c >>
O i—
CU T- C
C 4J O
JD O CO
i- Q. CO
3 -a
•«-» E 3
 C
c3 to t— <
to
4_>
i-
Q.
CU
^
r— i
(O
^j
C
c

3
t %
•»•'
•i^
4_>
to
C
>— i

O
•r—
t-
4->
O
CU
r™-
LU
C
o
to
•5
LU

M
to
cu
0
i.
Z3
o
to
(O
**
IO
0

. •
cu
o
i.
••i
o
co
                                                                                                                       r-|   OJ|
                                                                    3-20
                                                                                                            c/)
                                                                                                            CU

-------
lowest cost per kw was recorded in the early 1970's as sales reached their
peak.  However, the estimated figures for 1975 show an average cost per kw
to be around $75, the highest in 12 years.
     A survey of domestic utilities presented in Sawyer's 1975 Gas Turbine
Catalog gives some indication of future sales in this sector.  Their figures
showed no appreciable increase in demand for gas turbines as a percentage
of total generation mix between 1975 and 1985.  More than half of all utilities
anticipate that gas turbines will represent no more than 10 percent of
the future generation mix.  The primary reason given is the increased cost
of fuel and the unpredictability of oil and gas supplies.
     The analysis assumes that total generating capacity will grow at the
same rate as electricity demand (6.4 percent annually).  In addition, it
assumes that turbines will retain their current 8.25 percent share (39,300 MW)
of the total generating capacity which in 1974 was 475,897 MW.  Turbines
currently supply only 1.6 percent of the nation's power generation because
they are used primarily for peaking.  However, this ratio is not expected to
change.  Based on these assumptions, approximately 57,000 MW in generating
capacity would be attributable to gas turbines by 1980 and about 77,700 MW
by 1985.
     A 1974 study by Temple, Barker, and Sloan (TBS) provides an alternative
projection.  They predict that by 1985 "peakers" will make up only 7.9 percent
of the electric generation mix.  The TBS analysis concurs with EEA projections
for net generating output from gas turbines in 1985.  However, they also
predict that gas turbines will decrease their share of the generation mix
over the next two years and by 1985 will account for slightly less than
                                     3-21

-------
64,000 MW of generating capacity.   The Increase 1n  generating  capacity
would, therefore, be jnly 24,700 MW or 40 percent less  than  that Indicated
by EEA projections.
     The TBS study therefore Implies longer yearly  operating time for gas
turbine equipment.  However, this assumption contradicts  the 1975 user  survey
informtion on generation mix done by Sawyer's Catalog.
     To estimate unit sales from MW output data, it is  necessary to determine
the size distribution of the units being sold.  Energy  and Environmental
Analysis, Inc.,  (EEA) data indicates the following  trend  in  average gas turbine
unit sizes for utility applications over the past ten years:
CD
2
UJ
             90
             80
             70
             60
             50
             40
             30
             20
             10
                    66   67   68  69  70 71  72  73  74
                                YEARS
 "Units"  in  this  case  include combined cycle and tandem gas turbine configurations.
 Assuming that  the  average unit in the next ten years will be at least 80 MW,
 the  following  table indicates the range of market projections:
                            1980
                       Units      MW
                                             1985
                                        Units     MW
      EEA Assumption
      TBS Assumption
221
140
17,700
11,400
481
309
38,400
24,700
                                     3-22

-------
     3.1.2.4.3  011 and gas jndustry - Gas turbines 1n the oil and gas
Industry are used almost exclusively for oil and gas transmission and
extraction.  The gas turbines fit the needs of pipeline applications because
they are available in a variety of sizes and are suitable for semi-remote
control operation.  Most of these turbines run continuously, 8000 hours
per year.
     Solar has dominated the oil and gas transmission business 1n recent
years, especially in the smaller size ranges under 5,000 HP.  In the
5,000 - 20,000 HP range, or the upper end of gas and oil transmission
applications, foreign firms historically have been successful.  In 1970,
ten percent of the gas turbines purchased by the oil and gas Industry in
the United States were manufactured by foreign firms.  In recent years,
however, foreign sales have fallen off to virtually nothing.
     Gas turbines have a considerable capital cost advantage over recipro-
cating engines.  FPC data shown below indicates a  reciprocating engine costs 33
63 percent more than a comparable gas turbine.
Application
   $/Horsepower
Turbine    Reciprocating
Ratio of
Reciprocating
to Turbines
New transmission
  compressor stations       252
Additions to trans-
  mission compressor
  stations                  274
New field and storage
  compressor stations       353
Additions to field and
  storage compressor
  stations                  288
                337

                367

                565

                470
   1.33

   1.34

   1.60

   1.63
                                      3-23

-------
Diesels are used because they require less fuel  than turbines to
produce the same HP and consequently have lower operating costs.  Another
advantage of gas turbines over internal  combustion engines 1s that they
can be operated by remote control  and need not be Inspected very often.
Addition of NO  controls using water injection could eliminate this
              /\
advantage1 since the water treatment equipment would have to be inspected
periodically.  An informed industry source indicates, however, that most
users of pipeline transmission equipment see fit to have on-s1te Inspections
at least once a week.  This factor would seem to discount the idea that
installation of NO  controls are detrimental to the future competitive
                  rt
position of turbines.
     Calculation of future domestic gas turbine demand to 1985 in the oil
and gas extraction and transmission business will be based upon estimated
oil and gas production scenarios.   The Project Independence Blueprint
(PIB) provides two scenarios:  a $7 per barrel business-as-usual (BAD)
format and an $11 per barrel accelerated development (A-D) format.  The
PIB average annual compound growth rate estimates for the 1974 to 1985 period
are as follows:
                             011            Gas           Average
     BAU                    1.14%          4.30%           2.72%
     A-D                    6.03%          7.35%           6.7%
A range for potential future production can therefore be assumed to be
b cween 1.14 percent and 7.35 percent.
     EEA data indicates that total gas turbine horsepower installed through
1974 for the oil and gas industry, less electric power generation, amounted
                                      3-24

-------
to approximately 6,831,000 HP.  Using the average growth rate under
the business-as-usual format, the projected additional  demand for the
7-year period from 1974 to 1980 Inclusive would be 1,411,000 HP and for th«
12-year period through 1985 would be 2,592,000 HP.  Using the accelerated development
format, the projected additional 7-year demand would be 3,924,000 HP and
the 12-year demand would be 8,073,000 HP.  The growth rate 1n 1974 was
5.2 percent which would produce an estimate between the two already
calculated.
     The following graph indicates the fluctuations in  the average size
of gas turbines ordered for use by the oil  and gas Industry for extraction
or transmission purposes:
     6000
     5000
     4000
     3000
     2000
     1000
              65   66    67
68   69  70
   YEAR
71   72   73   74
Though the trend appears to be toward smaller Individual units, 1t 1s
reasonable to assume that the average size unit over the next 10 years will
be at least 3000 HP.  Using the 3000 HP figure, it is possible to calculate
the number of units to be sold.  Under the business-as-usual format,
approximately 470 units would be sold between 1974 and 1980, and 864
between 1974 and 1985.  Under the accelerated development format, there
appears to be a market for almost 1300 new units through 1980, and 2700 new
units through 1985.
                                    3-25

-------
     3.1.2.4.4   Stand-by electrlc generat1on - Small  electric generator sets
make up a considf ?ble number of all  gas turbine sales under 5,000 HP.
In the United States, the majority of these sets are  used for stand-by
electric generation, with the telephone companies constituting the principal
users.  In addition, some gas turbine stand-by units  are used by hospitals
and small municipalities.  These turbines are operated an average of between
75 and 200 hours per year.
     In a stand-by application, the annual  operating  time of the generating
equipment is typically under 100 hours.  The user requirements are quick
starts, low capital costs, ability to burn a variety  of fuels, high
reliability and easy maintenance.  The gas turbine meets these requirements.
     The American manufacturers producing gas turbines for stand-by applications
include Solar, Detroit Diesel Allison, Garrett and Lycoming.  Only five or six
different turbines are widely used in stand-by generation packages.
     The competition from reciprocating engines in this market is strong,
especially for applications under 1,000 HP.  A study  by market research
specialists Frost and Sullivan, Inc., estimates that for power generation
units under 3,000 kw (4,020 HP) diesels account for 68 percent of the market
and  by 1985 will account for 76 percent of the total  market.  They indicate
that the market in general will grow at a 12.6 percent annual rate with
diesels growing at a 15.3 percent annual rate.  Gasoline units in this
market will only grow at a 6.2 percent annual rate.
     The Frost and Sullivan study noted that trying to estimate the growth
rate of gas turbines in small generator sets was "one of the great enigmas
of our investigation".  Nevertheless, they predicted a conservative annual
growth rate for turbines of nearly 10 percent.
                                       3-26

-------
     The 1974 EEA data Indicates  that  the  total  horsepower capacity for
stand-by units 1s 1,256,842 HP  (or 937 MW).  Using a 10 percent annual
compounded growth rate, there would be an  Increase from 1974 to 1980  Inclusive of
1,193,000 HP (or 888 MW) and from 1974 to  1985Inclusive of 2,695,000  HP  (or 2000 MW),
     The average size stand-by  electric generator sets has remained
relatively small over the past  ten years as  the  following grapth Indicates:
      1500
    >i125C
    •H 1000
    |  750
    cv 50C
    *  250
              65   66    67   68  69   70   71    72  73   74
                                 YEAR

Assuming the average size unit will be approximately 1,000 HP over the  next
ten years, approximately 2,700 units will  be marketed for this application,
with 1,200 of them sold by 1980.
     3.1.2.4.5  Private Industry electric power generation - Electric power
generation by private industry was the third most Important use of stationary
gas turbine equipment in 1974 in terms of HP.   Large Industrial  complexes
and refining facilities consume considerable amounts of electricity.  Turbines
will run between 4,000 and 8,000 hours per year, depending on the application.
This category has been divided into electric generation by the oil and  gas
industry and electric generation by "other" private industries.
     A significant increase in generating capacity occurred in 1974 in  appli-
cations for the oil and gas industry.  The advantages of gas turbines over
                                     3-27

-------
steam electric plants in terms of reliability  and  flexibility  appear  to  give
gas turbines a very strong market position  in  this industry.
      A projection of future demand for gas turbine electric generating
equipment in the oil and gas industry can again  be based  on the  Project
Independence Blueprint high and low annual  growth  rates presented  earlier.
Under this scenario, the range would be between  1.14 percent and 7.35 percent.
The total generating capacity for the oil and  gas  industry in  1974 was
approximately 460 MW.  With a conservative  1.14  percent growth rate,  the
1980 total would be 500 MW and the 1985 total  would be 536 MW, or  an  increase
of 76 MW.  With a rapid growth rate of 7.35 percent, the  1985  total would
be 1003 MW or an increase of 543 MW from 1974  and  the 1980 total would be  704MW.
      The average size unit over the past ten  years has varied greatly with units
averaging 1 MW in 1972 and 1973 and 10 MW in 1974.  Estimating the potential
for gas turbine installations in this application  is therefore difficult.
However, using a 5 MW average size unit, the total installations through 1980
would be 8 units under the conservative growth rate and 49 with  the accelerated
development format, and through 1985 the number  of units  would range  from  15 to
108.
      For electrical generation from other  industrial  sources, an  estimation
of the annual growth rate of gas turbine applications can be based upon  the
estimated annual growth rate of the industrial sector.  Using  the  summer
1975 forecast of the United States economy  by  Data Resources,  Inc., a 5.979
percent, annual compounded growth rate can be established  from  the  industrial
production index.  EEA data indicates the total  industrial electrical generating
capacity in 1974 was approximately 1,900 MW.  Using a 6 percent  annual growth
rate, the 1985 generating capacity should reach  3,600 MW  or an increase  of
1,700 MW.
                                      3-28

-------
     The average size Installation over the past ten years has followed
the same highly variable pattern as did the oil and gas Industry, though
with a few units each year approaching the 20 MW level.  However, for
simplicity, the 5 MW unit will again be used as the average size unit
for future Installations Indicating approximately 340 new gas turbines
will be installed 1n the Industrial sector for electric power generation.
     3.1.2.4.6  Other industria] appllcations - Industrial applications for
gas turbines include various types of mechanical drive and air compression
equipment.  These applications reached their peak 1n the late 1960's and have
fallen to relative insignificance in recent years, both 1n terms of units
and horsepower installed.
     Gas turbines should meet continuing stiff competition from dlesel
engines for most of the smaller Industrial applications to which they are now
applied.  However, gas turbines do have several attractive attributes for
industrial drive applications.  These are their small size, high horsepower-
to-weight ratio, ability to burn a variety of fuels, reasonable reliability,
availability in self-contained packages, power without warm-up quality, lack
 of cooling water requirements,  vibration-free operation,  and low Installation
costs.  This flexibility is especially valuable where operating requirements
may run from 2,000 to 8,000 hours per year.
     Again, the estimated demand for Industrial gas turbine applications, less
electrical power generation, can be based upon the projected annual growth
of the industrial sector.  EEA data Indicates a total of 1,146,589 HP used
through 1974 for Industrial drive type gas turbine applications.  With the
6 percent annual growth rate, 2,175,000 HP of drive will come from gas turbines
                                     3-29

-------
by 1985 or an increase since 1974 of approximately  1,029,000  HP.   However,
this is probably a high estimate considering the few installations in  recent
years.
     Estimating the average size unit and the total  new units through 1985 is
of dubious value due to the great variety in size of installations.   There
were ""3 installations averaging 1,200 HP in 1972 and 3 installations  averaging
19,500 HP in 1974.
     3.1.2.4.7   Summary of market projections - Aggregation of the results  from
the five market sectors analyzed provide a profile  of the turbine market beyond
1974.  Tables 3-6 and 3-7 summarize the market projections through 1980 and through
1985, respectively.
     Utilities will continue to dominate the market with at least 80  percent
of the HP requirements.  Stand-by and oil and gas industry applications will
account for over 80 percent of unit sales.
     If the market projections are annualized, then the decline from  1974
order levels becomes more obvious.  Domestic orders for turbines in 1974
totalled 12.2 million horsepower and 434 units.  The 1974 through 1985 projection
produces an annual order rate of 3.8 - 6.0 million  HP and 383 - 574 units.
This represents a stabilization in unit sales which have been declining since
1970 but a drastic decline in HP sales.  Preliminary information about order
cancellations in 1974 seems to indicate that the downward adjustment  in
HP demand has already occurred although it is not reflected by existing data.
     The market projections also reflect tremendous uncertainty.  Capacity
and unit sales could vary by 75 percent and 90 percent, respectively, depending
on growth in power consumption and the future price of oil.  However, even
under the most optimistic assumptions, growth rates of the late 1960's
                                     3-30

-------
                                 Table 3-6
                        1974 - 1980 MARKET PROJECTIONS
Application
Utilities
011 and Gas
Stand-by
Private Electric Power
Generation
-- oil and gas
— other industry
Increase 1n
Capacity (M HP)
High Low
23.70
3.92
1.19

.33
1.06
15.30
1.41
1.19

.05
1.06
Increase 1n
Units
High Low
221 141
1300 470
1196 1190

49 8
159 159
Other Industrial Applications       .48        .48              N/A     N/A
TOTAL                             30.68      17.49             2919    1967
                                       3-31

-------
                                   Table  3-7
                         1974 -  1985 MARKET PROJECTIONS
Application
Increase 1n
Capacity (M HP)
High Low
Increase in
Units
High Low
Utilities                          15.50       33.10              481      309

Oil and Gas                         8.04        2.59             2700      860
Stand-by                            2.69        1.69             2690     2690

Private Electric Power
  Generation
  ~ oil and gas                     .73         .10              108       15
  -- other industry                 2.28        2.28              340      340
Other Industrial Applications       1.02        1.02              N/A      N/A

TOTAL                              66.26       41.78             6319     4214
                                       3-32

-------
and early 1970's are not expected to reappear unless new turbine developments
occur or new applications are discovered.   Table 3-8 presents a summary of the
methodology used 1n calculating the growth projections.
3.1.2.5  Balance of Trade Position  -  Over the last five years, gas turbines
have Improved their already positive balance of trade position.  Exports
have increased considerably and have helped to make up for the reduction 1n
domestic sales caused by increased fuel prices.  Imports, on a direct sales
basis, have fallen to relative unimportance.
     The export market for U. S. manufactured gas turbines was about 5.46
million HP in 1974, or about 30 percent of total U. S. gas turbine output.
Exports for the last five years were as follows, as Indicated by EEA data:
                      1970      1971      1972     1973      1974
     Units             194       257       198      291        372
     HP (000)          914.4    1572.5    1837.4   7079.6    5458.4
     Although growth has not been steady, the total HP sold has more than
quadrupled from 1970 to 1974, while the number of units sold has Increased
by about 90 percent.  The difference in the magnitude of the HP and unit
increases reflects the trend toward larger Individual units used in
electric power generation.
     The U. S. export market for gas turbines 1s dominated by five firms:
General Electric, Avco Lycoming, Solar, Turbo Power and Marine, and Westinghouse.
These five firms accounted for about 87 percent of the units exported 1n
1970 and about 90 percent of the units exported 1n 1974.  Their market
control appears independent of yearly export fluctuations.
     The three firms shown below are the largest exporters in terms of HP and
unit sales:
                                      3-33

-------
                                  Table 3-8A
                         METHODOLOGY  FOR MARKET PROJECTION

The above calculations were done using the  following methodology:
(1)  Capacity in place was determined for 1972 or  1974;
(2)  A range of annual growth rates was determined for  the  each of  the
     sectors using turbines;
(3)  The increase in capacity was calculated using the  formula:
                          AX  -  XQ  (Hr)n  -  XQ
     where AX  =  increased turbine capacity between base year and  projection year
            XQ =  capacity in place in the  base year
            r  =  annual demand growth rate
            n  =  number of years between base year and the projection year
(4)  An average unit size (U ) was determined based on  historical data on
     turbine sizes;
(5)  The number of new turbines was calculated using the formula:
            number of new units  = AX
                                   \
(6)  Because a turbine will last a minimum  of 20 years  and  because  over  90
     percent of turbine capacity was  installed after 1965 (see Figure 3-3),
     the replacement market was assumed to  be a negligible  portion  of new
     turbine sales.
                                      3-34

-------
% of
Total HP
40
8.5
35
% Of
Total Units
23
45
5
Average Unit
Size (HP)
27,276
2,800
94,260
General Elec.
Solar
Westinghouse
     GE units are used primarily for electric utility power generation.
Solar supplies the smaller gas compression, generator drive, and oil and water
pump markets.  Westinghouse units are used primarily for generator drive.
Of the other two firms, Avco Lycoming Division has exported smaller units
used mostly for marine purposes whereas Turbo Power and Marine has exported
larger units used mostly for peak load requirements.  Firms dealing 1n larger
 units seem to have experienced a relatively stable growth  in  their exports
while the exports of those dealing in smaller units have fluctuated from
year to year along with the total export market.
     The Department of Commerce has placed the dollar value ($ million)  of
turbines shipped as follows:
                          1970       1971       1972      1973      1974
Gas Turbine Generator
  Sets Assembled or Un-
  assembled               $13.6      $30.6      $21.9     $98.2     $150.6
Gas Turbines and Parts
  for Mechanical Drive,
   NEC!/                   $51.0      $71.1       $86.9    $137.0     $153.5
TOTAL                     $64.6     $101.7     $108.8    $263.2     $304.1
According to these figures, the dollar value of gas turbines, gas  turbine
generator sets, and parts for mechanical drive shipped abroad has more than
quadrupled since 1970.  Turbine shipments now contribute over $300 million
to the balance of payments and are growing rapidly.
]_/  NEC =  Not elsewhere classified
                                      3-35

-------
40
885.8
40
408.7
18
275.9
8
13.1
3
29.5
     During this same period, Imports of gas turbines  Into the United States
have been steadily declining.  The total HP Imported was only about 3 percent
of the 1970 total.  The data compiled by EEA, which does not Include foreign
manufactured equipment sold to American packagers, 1s  as follows:
                               1970        1971      1972       1973       1974
     Units
     HP (00)
     Doubtless, the devaluation of the U. S. dollar influenced the increasing
U. S. exports and reduced imports of gas turbines.  The predominance of U. S.
exports seems likely to continue.  Department of Commerce figures concerning
the shipments of gas turbines and parts for mechanical drive  (not elsewhere
classified) which were cited above show that $110,847,457 worth of merchandise
in this grouping have already been shipped 1n the first half  of 1975.  If this
rate of shipments continues, the 1975 dollar value exported will surpass 1974.
3.2  PROCESSES AND THEIR EMISSIONS
3.2.1  Processes
3.2.1.1  The Simple Cycle Gas Turbine   -  A gas turbine is a  rotary engine
which consists typically of  one or more  compressor stages, one or more
combustion chambers where liquid or  gaseous fuel  1s burned,  and one or more  turbines
to drive the compressor(s).  For stationary applications, a  power turbine is
included to drive the load.  There are  various configurations of these basic
gas turbine components.  Figure 3-5  presents a block  diagram showing  a typical
s.mple cycle gas  turbine.  Figure 3-6 presents a  cut-away view of a dual spool
gas turbine with  can-annular combustors  and a free power  turbine.   It is called
a dual spool turbine because the low pressure compressor  and the high pressure
compressor stages consist of two mechanically independent rotor systems and  the
                                      3-36

-------
     g
                                    J-
                                    (O
                                    CT>

                                    O)

                                    O
                                    >,
                                    O

                                    OJ
                                    (O
                                    O
                                   LO
                                    I
                                   ff)

                                    
-------
                 -GAS GENERATOR SECTION
                                              FREE TURBINE SECTION-
                                                                       EXHAUST GAS
                                                                     r- DISCHARGE -,
AIR INLET
 SECTION
LOW PRESSURE
 COMPRESSOR
          HIGH PRESSURE
           COMPRESSOR
    ROTOR STAGES
       STATORSTAGES
      N1 ROTOR
                                                                          POWER SHAFT
                                                                           COUPLING
              POWER
             TURBINES
                                COMBUSTION
                                 CHAMBER
FREE TURBINE
   INLET
 EXHAUST .
COLLECTOR
                 Figure 3-6. Cut-away view of a typical simple cycle gas turbine.
                                          3-38

-------
power turbine is "free" because it is not mechanically connected to either
compressor rotor.  The combustion chambers "cans" are arranged in an
annular configuration around the gas turbine axis, hence the term can
annular.  Figure 3-7 shows a two shaft turbine with an annular combustor.
The compressor assembly and compressor drive turbines are mechanically
attached to the same shaft while the free power turbine is mounted on a
separate shaft which connects to the load.  This turbine is powered by a
single annular combustor in contrast to the multiple can combustors shown
in Figure 3-6.  Gas turbines are also produced in a single shaft configuration
wherein the compressors and the turbines to drive the compressors and the  load
are all mechanically attached to the same shaft.  The turbine shown in Figure
3-8 uses a single external combustion chamber which is about 10 feet in
diameter, 21 feet in height, and weighs 10 tons.  The exact configuration
of a turbine (type of comhustor(s) used, dual spool, two shaft, single
shaft, etc.) depends to some extent on the manufacturer's past experience
and preference and also on the specific market application (e.g., peaking
turbine used by utilities, gas pumpers, etc.).
       Gas turbines are started by using electric motors, diesel engines,  com-
pressed air or another energy source to rotate the compressor in order to  provide
sufficient air under pressure to support combustion in the combustors.  Fuel
is then introduced into the combustors and burned to produce hot gas which
expands across the first set(s) of turbine blades, thus providing the driving
force to continue rotating the compressor(s) mounted on the same shaft. The
hot gas is further expanded across the power turbine blades which convert
some of the remaining energy in the gases to output shaft power for driving
                                     3-39

-------
Figure  3-7  Cut-away view of a simple  cycle, two  shaft
             turbine with an  annular combustor.
                                                               OUTPUT SHAFT
                                                            EXHAUST COLLECTOR


                                                  TURBINE ASSEMBLY
                                       COMBUSTOR
                 COMPRESSOR ASSEMBLY
                                  3-40

-------
                                                          COMBUSTION
                                                           CHAMBER
Figure 3-8. Single shaft gas turbine with an external combustion chamber.
                           3-41

-------
pumps, electrical generators, etc.  The heat energy remaining in the gases
after expansion across the power turbine can then be further utilized in
steam boilers or other heat exchangers.
3.2.1.2  The Regeneratiye Cycle Gas Turbi ne  -  The hot combustion gases
from the simple cycle gas turbine (described in Section 3.2.1) exit to
c^nosphere at exhaust gas temperatures ranging from about 800 to 1100 F.
The regenerative cycle gas turbine is essentially a simple cycle gas turbine
with an added heat exchanger as shown in Figure 3-9.  In the regenerative
cycle gas turbine, thermal energy from the exhaust gases is recovered and
utilized to preheat the compressor exit air.  Since less fuel is required to
heat the air to the design turbine inlet temperature, increasing the temperature
of the air injected into the combustors decreases the fuel requirements of the
turbine and improves the overall cycle efficiency.
3.2.1.3  The Combjned Cycle Gas Turb jne  -  The combined cycle gas turbine also
recovers waste heat from the turbine exhaust gases.  It is essentially a
simple cycle gas turbine with the hot exhaust gases vented to a waste heat
recovery boiler as shown in Figure 3-10.  Steam generated in the boiler can
be used in a number of ways including generation of electricity via conventional
steam turbines.  Some waste heat recovery boilers are designed to permit
generation of additional steam by burning conventional fuels in a firebox
to supplement the heat from the hot exhaust gases.  Such systems are referred
to is supplementary-fired combined cycle gas turbines.
3.2.2  Emissions from Stationary Gas Turbines
     The pollutants emitted from gas turbines are those common to all combustion
sources, NO , HC, CO, S09, particulates, and visible emissions.  The mass
           X\            Cm
emissions from stationary gas turbines will differ depending on several
                                       3-42

-------
                                                  a
                                                  <:
                                                  o
       rmr
M31\/3H3dd
r^
i

inrtnrtnnfi) o o o o


	 r^ — i —
	 1 UJ
cc
a.

                                                  cc
                                                  o
                                                  CO
                                                  cx>
                                                  LU
                                                  CC
                                                  a.

                                                  o
                                                  o
                                                                                OJ
                                                                                c
                                                                                •I—
                                                                                -Q
                                                                                S-
                                                                                ra
                                                                                en
                                                                                0)
                                                                               r—
                                                                                u
                                                                                0)
                                                                                (O

                                                                                (U


                                                                                I
                                                                                0)
                                                                                to
                                                                                o
                                                                               (Tl
                                                                                I
                                                                                O)
                a.

                o
                o
                                     a:
                                     
-------
a

o
                                                                              c.
                                                                              C

                                                                             .Q
                                                                              i.
                                                                              =3
                                                                              >
                                                                              o

                                                                             "O
                                                                              O)

                                                                             •I—
                                                                             ^3


                                                                              O
                                                                             (O
                                                                             o
                                                                              i
                                                                             oo
                                                                             O)
                  3-44

-------
variables which are discussed 1n greater detail 1n the following section.
Examples of some of the variables are:  turbine firing temperature, turbine
pressure ratio, turbine load, combustor design, and atmospheric conditions.
3.2.2.1  Simple Cycle and Combined Cycle Gas Turbines  -  Mass emissions from
identical gas turbines used 1n the simple cycle and combined cycle (unfired
boiler) configurations are equivalent and, therefore, only emissions from
simple cycle machines will be discussed.  If the combined cycle gas turbine
uses a boiler with supplementary firing, the total emissions to atmosphere
will be increased by pollutants generated in the boiler.  Table 3-8 shows typical
visible, CO, SO,,, and NOX emissions for turbines of various sizes  manufacturered
by several companies when fired with both natural  gas'amPMquid fuels.   The table
shows that for turbines operating at base load visible emissions are generally
less than 10 percent opacity, NO  emissions range from 0.94 Ib/hr to 1578 Ib/hr
                                j\
for turbines having outputs of 0.16 MW and 87.8 MW, respectively, and CO emissions
range from 0 Ib/hr to 47.2 Ib/hr for turbines having outputs of 60 MW and
51.7 MW, respectively.
     Sulfur dioxide emissions are a function of the efficiency of the gas
turbine and the sulfur content of the fuel, since virtually all fuel sulfur
                      789
is converted to oxide. '  '    Gas turbines fired with premium low-sulfur
fuels have low S02 emissions.  A 33 MW gas turbine will produce about 36 pounds
of S02 per hour when operated on 0.1 percent sulfur fuel, but emissions from
this same turbine will increase to 690 pounds per hour when fired with 1.9 per-
cent sulfur fuel.  Table 3-8 shows that for turbines of various sizes, manu-
factured by several companies and fired with liquid fuels ranging from 0.01
percent to 0.5 percent sulfur by weight, SO,, emissions range from 0 Ib/hr to
540 Ib/hr for turbines having outputs of 0.8 MW and 88 MW, respectively.
                                       3-45

-------
                                                                      C-'S-
                                                                      s-I
                                                                                                                               1
                                                                                                                               5
-o IB •—

U VI 1o
.    .
                                                                                                      .SSB
                                                                    -===      =

                                                                    1^
                                                                                                     fas
                                                                                                     a en o
                                                                                                      ••- O. K
                                                                                                      ••- O.
                                                                                                     *r ** ai
                                                                                             *• o
                                                                                             °
                                                                                             • U4

                                                                                             !»•
                                                                                               fv
                                                                                             ^St
                                                                                             tS>
                                                                                             2c
                                                                                              1 °
                    oo   o o      oo   do      odd  d <— t
                    CM mo --r-~    o-^foor—in    ocMtoSo^eMO   Sc?So2ScochccS'^SS

                    otM   o O      o r-^   dr--      r-^roio  ^WKi   r-^t>^«s   ^cnindeM^^roin
                                                                                                                  Or— 10COO  Vr—
                    °(V^OO.°^    O r-tC o r-> r~-    o eC r-CM c InS in    10 r-»   o ^-     ^T

                    oo   C3O      OP—*   df*i      ddr-^  (^dr^    di-^M   ^rnininior-

                                                            *~          ,—     ^j O->   ,_ <-
                                                  O C 0  O C o    O O O   OOOOOOOO
                    d   do      dd   d r-^     od    oo        o     o
                                                        Or-'        O     O
                                                 8S     tR
                                                 3^J     ^J m
                                                                    Or-       10     O
5B
 CM^
Sc
                                                CM 00     O g


                                                esj O     CM CM
                    in   i-^ r--     o *  q> O     >n o     •» °
                    rMro
                    o  o

                    V  V
                                    CO       O    C3


                                    V     V       V    V
                        in        m
                                                                    »    s.       »     a.
                                                                    r-T    o       eo     o
                     r- r- in -J-      CO

                     o o o o      ^
                                                                           s-s
                                                                                             a    2
                                                                                                   te     fe
                                                 0       0
                                                                           CD      CO
                                                                     if?            B
                                                                                                                                               3-46

-------
                      TABLE 3-8B (Continued)
  NOTES
A.  NOV and CO emissions were calculated using one of the following
      X
    formulas:
     1)  ER = 1.552 X 10"7  (Ec) MQ
         where:  ER = pollutant emission rate (#/hr)
                 EC =     "     concentration (ppm), dry basis
                          (not corrected to 15% 02)
                 M = pollutant molecular weight
                          NOX (as N02) = 46
                          CO = 28
                 Q = stack flow (SCFM), at 29.9"Hg and 70°F

     2)  ER = EpPh
         where:  ER = pollutant emission rate (#/hr)
                 E  =     "   mass  "     "   (#/1000 hp-hr)
                 Ph = power output (hp)
     3)  ER = Eh (Hf) Wf
         where:  Eh = pollutant mass emissions rate (I/million Btu)
                 Hf = lower heating value of fuel (assumed 18,400 Btu/#)
                 W  = fuel usage rate (#/hr)

B.  Heat rates, when not provided directly, were calculated as follows:
     1 )  HR = SFC (Hf )
         where:  HR = Heat Input required per unit of work ( Btu )
                                                            kw-hr
                 SFC = specific fuel consumption (l/kw-hr)
                 H  = lower heating value of fuel (Btu/#)
     2)   HR =  WfHf
         where:  Wf = fuel usage (#/hr)
                 H-. = lower heating value of fuel (Btu/#)
                 P.  = power output (kw)
                                 3-47

-------
                        TABLE  3-8B  (Continued)
C.  Sulfur to S02 conversion is assumed to be 100%, using the following
D.
E.
    formula:
        ER  a
                   HR  (S)  p.
        where:  ER = S02 emission rate (#/hr)
                M = molecular weight
                S = fuel sulfur content (#S/#fuel)
                HR = gas turbine heat rate (Btu/kwh)
                P.  = power output (kw)

    For the instances where emissions were measured at othpr than
    rated base load, the actual power output during the tests is
     indicated by the presence  of brackets.

    1)  S.C.  refers to a simple cycle gas turbine
    2)  R.C.  refers to  a regenerative cycle  gas turbine
    3)  CCUF  refers to a combined cycle plant with an unfired heat.
        recovery system generator.
    4)  N.G.  refers to natural gas
    5)  DF-1  refers to distillate fuel #1
    6)  DF-2  refers to distillate fuel #2
                                3-48

-------
     Unburned hydrocarbon emissions from gas turbines are due to vaporized
unburned fuel or partially burned products which escape the combustion
reaction zone and are emitted in the exhaust.  Since combustion efficiencies
of gas turbines operating at base load normally exceed 99 percent, hydro-
carbon emissions at base load are not substantial and may range from 1 to
5 ppm total hydrocarbons measured as methane or hexane (1 ppm is approximately
                                      11, 12, 13, 14
equal to 3 Ib/hr for a 30 MW turbine).                Ambient hydrocarbon
levels during many of these tests for hydrocarbon emissions were also measured
as 1 - 5 ppm.
     Particulate emissions from gas turbines consist of ash from the fuel and
particulates of carbon and hydrocarbons resulting from incomplete combustion.
Depending on the size of  the turbine and the fuel burned, particulate
emissions range from 1 Ib/hr to over 40 Ib/hr and vary from 0.002 gr/scf
to 0.10 gr/scf.  '   '   'For example, particulate emissions from a turbine
operating at 20 MW and burning natural gas were 4.8 Ib/hr while those from a
turbine operating at 52 MW and burning #2 oil were about 41 Ib/hr.  Somewhat
higher levels of particulate emissions may result from gas turbines fired with
crude or residual fuels, although some data indicate that particulate emissions
when burning these fuels are actually less.  When considering these data on
particulate emissions, one should remember that the high flow velocities,
extreme gas turbulence from the ducting configuration and sound baffles and
the difficulty of sampling with very long probes in the environment of the
exhaust gases results in particulate measurements of questionable accuracy
and high variability.21' 22' 23
     Visible emissions or smoke emissions from gas turbines may be caused by
only a small portion of the total particulate emissions since they are
comprised of the extremely small and finely dividied particulate matter.
                                    3-49

-------
Various studies have Indicated that the major contributor to visible
emissions are the p?~t1culates consisting of low density agglomerates of
carbon which are generally less than 1  micron in size and of the same
order of magnitude as the wave length of visible light thus producing a
                                               24  25  w-
greater visual effect than larger particulates.  '     Table 3-8 shows
that i  isible emissions from stationary gas turbines operating at base load
are generally less than 10 percent opacity.
3.2.2.2  Regeneratiye Cycle Gas Turbines  -  The kinds of pollutants emitted
by regenerative cycle gas turbines are identical to those listed in Section
3.2.2.1 for simply cycle and combined cycle turbines.  Table 3-8 shows NO
                                                                         X
CO and SO- emissions for a General Electric MS 7001 B  simple cycle machine
and a MS 7001 B regenerative cycle machine.  In general, the following
observations can be made for the two machines when operating at equal outputs:
(1) CO emission levels are essentially the same for both machines; (2) NO
                                                                         A
emissions from the regenerative machine are greater by more than a factor of two;
(3) S02 emission levels are less from the regenerative machine and are proportional
to the relative efficiencies of the two machines.-'
     No data are available for particulate emissions from regenerative cycle
gas turbines.  Since particulate emissions occur from the fuel ash and from
the carbon and hydrocarbon particles formed by incomplete combustion, the
regenerative cycle machine should have lower particulate emissions than
an equivalent simple cycle machine.  This is a logical conclusion since the
-   This follows from the fact that all sulfur in the fuel is converted to S0?
and the higher efficiency regenerative cycle machine requires less fuel for
an equivalent output.
                                     3-50

-------

-------

-------
                            js fuel (and, hence, less *sh 1s form«H) p*»r
                          .oustion chambers for both machines are Identical.
                       jns from regenerative cycle machines are of the same
                                              26
                     om simple cycle machines.,
                  chat Inf1uence Emjss1pns from Statlonary Gas Turbjnes
                jits emitted from stationary gas turbines are CO, HC, particulates,
SOp           visible emissions (as detailed in Section 3.2.1).  There are
several ...tors which influence the amounts of these pollutants which are
emitted.  These factors are discussed, by pollutant, in the following sections.
SOp emissions are strictly a function of the sulfur content of the fuel and,
therefore, are not discussed further.
3.2.3.1  Operation of a Typical "Can" Combustor  -  The formation of pollutants
occurs during the process of combustion.  In order to understand the para-
meters that influence emissions from stationary gas turbines, one must understand
the basic operation and functions of the combustor.  A schematic of a conventional
                                                       27
style "can type" combustor is presented in Figure 3-11.    (The general
description of the combustion process which follows will also be applicable
to an annular type combustor.)  Fuel is commonly introduced to the combustor
can by the use of one or more nozzles which may use air or pressure atomization.
The combustion air, cooling air, and dilution air are supplied to the
combustor from the discharge of the compressor.  The mixture of combustion
air and fuel is ignited by an automatically retracting spark plug as shown
in Figure 3-11 and when multiple combustors are used, the flame is propagated
between combustor cans by "crossfire tubes".  The combustion of the fuel is
completed in the "reaction zone" where gas temperatures exceed 3040°F.   ' 29» 30
                                       3-51

-------
 RETRACTABLE
  SPARKPLUG
                                  COOLING
                                   AIR
                       DILUTING
                         AIR
                                                             TRANSITION
                                                                PIECE
CROSSFIRE
  TUBE
      ^PRIMARY ZONEH
INTERMEDIATE
    ZONE
  COMPRESSOR
 DISCHARGE AIR

-^-DILUTION ZONE-
       Figure 3-11.  Can type combustion system tongas turbines.27
                                                                         TURBINE
                                                                         NOZZLE
                                        3-52

-------

-------

-------
 The reaction zone consists of two subzones  -  the primary zone and the
 intermediate zone which are shown in Figure 3-11.  The boundaries between
 zones are not well defined and reactions of one zone may extend Into the
 subsequent zone.
     Efficient combustion of the fuel requires proper fuel atomization,
 proper air to fuel ratios, and mixing, and sufficient time (residence time)
 within the reaction zone of the combustor for the chemical kinetics of
 the reaction between the fuel and oxygen to be completed.  The combustion
 process is commonly stabilized in the primary zone of the combustor by
 aerodynamic recirculation of the combustion air entering the first row of
 holes in the combustor.  The portion of air that flows upstream toward the
 fuel nozzle (See Figure 3-11) is preheated as it mixes with the products of
 combustion and thus provides continuous ignition of the fresh charges of fuel.
 This continuous ignition of injected fuel by the preheated recirculation
 gases is important to provide stable combustion over a wide range of combustor
                                        31  32  33
 inlet conditions and air to fuel ratios.  '   '     Completion of the combustion
 reactions occurs in the intermediate zone.  Cooling air is introduced in the
 intermediate zone to provide a boundary layer along the Inside wall of the
 combustion chamber and to remove heat from the wall by convection.  Dilution
 air, admitted in the dilution zone, then mixes with the products of combustion
 to reduce qas temperatures to the design turbine Inlet temperature which can
approach 1950°F in present day machines operating at base load.
     Satisfactory combustor operation requires attainment of the followina
                                                             OC  OC
 objectives (not considering the reduction of air pollutants):  '
     a.  efficient combustion of the fuel (or fuels)
     b.  low pressure drops across the combustor (since any pressure drops
 are parasitic and result in reduced cycle efficiencies)
                                    3-53

-------
     c.  stability of operation over a wide range of fuel  to air ratios
(since fuel to air ratio varies proportionately with variations 1n
turbine load, this ratio will  therefore change over a wide range each
time the turbine 1s started and taken to full  load operation)
     d.  dependable Ignition
     e.  low component maintenance and long life
     f.  a controlled temperature distribution at the turbine inlet such that
temperature profiles do not exist which will result in failure of the turbine
buckets.
3.2.3.2  Factors Which Affect Particulate and Visible Emissions from Stationary
Gas Turbines  -
     3.2.3.2.1  Particulate emissions - Particulate emissions consist of ash
carbonaceous agglomerates and HC.  The ash content 1s primarily a function of
the fuel.  Fuels containing high ash and vanadium contents will result in
higher particulate emission rates than light distillate fuels.  Also,
inhibitors added to the fuel to retard vanadium corrosion of the turbine com-
                                            37  38
ponents will increase particulate emissions.  '
     Under some state and local regulations, unburned hydrocarbons (discussed
in Section 3.2.2.3) are considered as particulate matter if they are exhausted
                                                                39
as liquid droplets or allowed to condense in the sampling train.    However,
this should not add a significant amount to the emission rate except at turbine
                                  40  41
start-up and at low turbine loads.   '     All these particulate emissions
contribute to particulate loading.
     3.2.3.2.2  Visible emissions - Smoke - Although literature references
are somewhat contradictory, it is generally agreed that the major contribution
to visible emissions result from gas turbines when particles of carbonaceous
                                        3-54

-------
material formed during the combustion process agglomerate to sizes that
are in the range of the wavelength of visible light (0.45 to 0.65 micro-
meters) and form smoke.  These carbonaceous particles are caused by the
incomplete combustion of hydrocarbon fuels and the thermal cracking of fuel
                                                                 42  43  44
in hot regions of the combustor with locally insufficient oxygen.  '
A plume discoloration can occur under certain circumstances due to conversion
of the emissions of NO to NO^.  Factors which affect visible emissions are
discussed below.
          3.2.3.2.2.1  Fuels  -  Some fuels may form more smoke or visible
emissions than others if compensatory design changes are not incorporated
in the combustion system.  Generally speaking, visible emissions caused by
these carbonaceous materials are not experienced when gaseous fuels are used
as shown in Figure 3-12, which shows a smoke spot reading of zero for a turbine
burning natural gas (smoke spot number increases with increasing visible
emissions).  For liquid fuels, the paraffinic saturated fuels tend to "smoke"
less than the aromatic or unsaturated fuels and this smoking tendency is
related to the chemical bond energies necessary to completely consume the
fuel.46' 47
     Fuel hydrogen content and residual carbon content also affect visible
emissions.  A reduction in hydrogen content or an increase in residual
carbon, or both, can increase visible emissions.    Figure 3-13 shows the
effect of fuel hydrogen content on smoke for various turbine operating loads.
Figure 3-14 shows the effect of fuel hydrogen content on smoke for two
designs of gas turbine combustors, one specifically designed to reduce visible
emissions.  Figure 3-15 presents visible emissions versus the firing temperature
of the gas turbine when burning a typical east coast residual fuel.  All these
                                      3-55

-------
   ;ure 3-12.
             45
Visible  Emissions vs. Load for
a Turbine Using Various  Fuels
              3MCKE SPOT NU>eeRU3NC\«>*IOU3 FUELS
SMOKE SPOT"
ASTM 0-2156
                             ©•ZOSTILLATE FUEL
                             ANATURM.6AS
                 10     is"
                  LO4DIMW1
            20
                        3-56

-------
 Figure  3-13.49  Visible Emissions  vs.  Load for Two  Fuels
                 with  Different Hydrogen Contents
                 (GE MS-7001 and MS-9001  Turbines)
             SMOKF NUMBER

          BACHARACH VON BRAND
                  100 T
                   95
                   90
                  12.«% HYDROGEN
                                   11.0% HYDROGEN
                    10
                        20
                            30   40   50
                              LOAD (MW)
                                        60
                                            70
 Figure 3-14:
             50
Visible  Emissions vs.  Fuel  Hydrogen Content
for Two  Combustor Designs  (GE MS-7001B  and
MS-9001  Turbines)
             SMOKE NUMBER

          BACHARACH VON BRAND
                   100
                   96
                                      LOW SMOKE LINER
                          11.0     120     13.0
                        FUEL HYDROGEN CONTENT * BY WEIGHT
                                               140
Figure 3-15.
             51
Visible  Emissions vs.  Firing Temperature
(Load) for a GE MS-5001  Turbine Burning
Residual  Fuel
            SMOKE NUMBER
MARA
1
2
3
4
CH VON E
100
95
-
90
'
4(
RAND
/ 	 —
/
III II II
X) 500 600 700 800 900 1000 11
                           FIRING TEMPERATURE °C
  NOTE:  Visible  emissions decrease with increasing Von Brand
         number and decreasing  Bacharach number.
                          3-57

-------
turbines (Figures 3-13, 3-14, and 3-15)  were equipped with  combustors to
reduce visible emissions except for the  one labeled "orlg.  Uner"  1n
Figure 3-14.   With the one exception,  emissions were not 1n the  visible
      52
range.
           3.2.3.2.2.2   The effect of ambient temperatures on visible emjssions
The effect of ambient inlet temperature  variations  on the visible  emissions from
gas turbines  is shown for one engine in  Figure 3-16.  Reported results of
the effect of inlet temperature on visible emissions have been contradictory
but the experience of at least one other manufacturer indicates  a  slight
                                             54
increase in smoke with decreased temperature.    This increase is  believed
due to the increased quenching effects of the main  combustion air  and cooling
air at the combustor side walls.
           3.2.3.2.2.3  Other factors  -  The other factors which  affect visible
emissions are combustor design parameters which primarily relate to the
fuel to air ratios and mixing of the fuel and air.   '   '   '     Carbon can
be formed in the primary zone of the combustor if local areas of fuel rich
mixtures exist.  Combustor operating pressure can affect visible emissions
as can be observed from Figures 3-12,  3-13, and 3-14 (since combustor pressures
vary with operating load and firing temperature).  Possible explanations for
the effect of pressure level are changes in fuel droplet evaporation
temperature with pressure, differences in oxidant diffusion rates  with
pressure and changes in the spray parameters of the combustion  chamber fuel
                     59
nozzle with pressure.    It is generally agreed, however, that  the most
important considerations for reducing visible emissions involve  designs which
provide leaner fuel to air ratios (less  fuel; more air), minimize the number
of localized fuel rich areas, provide effective fuel atomization and mixing
                                     3-58

-------
Figure 3-16.
            53
Visible Emissions  vs.  Shaft Horsepower for Various
Compressor  Inlet Temperatures (GM, 501K Industrial
Turbine, Constant  Speed,  JP-5 Fuel)
         EM
         SMOU
         Hum*
        ttinaurr)
                20-
  10
                        1000
 2003
Stun
                       3000
WOO
  NOTE:   Visible emissions increase with  increasing EPA
          smoke number.
                             3-59

-------
with air and result in sufficient lean regions within the combustor for
   .   .      ,.59, 60, 61, 62, 63
smoke burnout.      '
3.2.3.3  FactorswVch Affect Emissions of Hydrocarbons and Carbon Monoxide _-
     3.2.3.3.1  Effect of combustion effici ency and operati on at 1pw turbj ne
loads - Incomplete combustion is the principal cause of emissions of
hydrocarbons (HC) and carbon monoxide (CO).   '   *      Since gas turbine are
typi"?lly designed for optimum efficiency at full  load, they normally operate
in excess of 99 percent combustion efficiency at full load.  For gas turbines
used in aircraft applications, this efficiency typically drops to the 90 to 95
percent range for operation at idle or low power conditions.  *     A similar
efficiency drop occurs in gas turbines used for stationary applications.  Because
of this drop in efficiency, emissions of HC and CO from the turbines will be higher
for turbine start-up and operation at low loads and will be a minimum at full load
operations.  Figure 3-17 shows HC emissions versus fuel to air ratio, and Figure
3-18 shows CO emissions versus firing temperature.   Since fuel to air ratio and
firing temperature are proportional to gas turbine load, Figures 3-17 and 3-18
demonstrate the trend for HC and CO emissions to increase with decreasing turbine
load - becoming an exponential increase at low loads.
     The low combustion efficiency (and, hence, higher HC and CO emissions)
experienced in gas turbines operating at low loads is due to the poor burning
conditions.71' 72' 73' 74  Low combustion inlet air temperatures cause quenching
to occur thus terminating combustion before completion, and the low fuel to
air ratios (fuel lean combustion) result in low equivalence ratios -  in the

- Equivalence ratio  (0) is defined as the ratio of the actual fuel to air ratio
to the stoichiometric fuel to air ratio.  For "fuel rich" combustion (excess
fuel) 0 exceeds 1.0 while for "fuel lean" combustion (excess air), 0 is less
than 1.0.
                                          3-60

-------
Figure 3-17
             69
Hydrocarbon Emissions  vs.  Fuel-to-A1r  Ratio
(Load)  for the GE MS-7001B and MS-5001N Turbines
When Burning 011 Fuel
            HC
          EMISSION
          INDEX -
          GR/KQR
           FUEL
                          0010
                          FUEL-AIR RATIO
                                           0.020
Figure 3-18
                70
   Carbon  Monoxide Emissions vs. Firing
   Temperatures (load)  for GE Gas Turbines
   Burning 011 Fuel
          CO - PPM
           (vou
                     500   700   900  1100
                      FIRING TEMP - «c
                             3-61

-------
primary zone of the combustor thus reducing burning intensity.   The low
fuel and air flows at low power operation result in poor fuel  atomization
and distribution.
     Figure 3-19 shows the relationships between idle power combustion
efficiency and CO and HC emissions and represents data from tests performed
on several aircraft-type gas turbines manufactured by General  Electric.
The specific ratio of HC and CO emissions from gas turbines operating at
low power conditions (and, therefore, lower efficiency) also varies with
specific design features of the engine combustor and fuel injection
systems.  As shown in Figure 3-19, the ratio of CO to HC emissions increases
as the efficiency level increases.  This is consistent with the chemical
kinetics of combustion reactions which show that HC compounds are consumed
faster than CO with the result that as qas turbine efficiency is increased, any
                                                                                 •jc -i->
remaininq non-equilibrium products of combustion will tend to exist mainly as CO.   '
The data from tests of these same engines also show that the levels of CO and
HC emissions from gas turbines with high compression ratios (ratio of
compressor inlet air pressure to compressor outlet or combustor inlet pressure)
are less than CO and HC emissions from turbines with lower compression
ratios.  This relationship is demonstrated in Figure 3-20, and one can conclude
that the increased combustor inlet temperatures and pressures due to higher
engine compression ratios generally result in improved combustion efficiency
at idle or low power conditions.
     3.2.3.3.2   Effects of ambjen t cgnditions  -  Figure 3-21 demonstrates the
effect of variations in ambient temperatures on CO and HC emissions from the
501K industrial  gas turbine produced by Detroit Diesel Allison, Division of
General Motors.  This figure shows that increases in ambient temperature result
                                       3-62

-------
                     Pounds Of H/C's  Per  1000 Pounds  Of  Fuel
                                                              Is)
O


3
CL
05
O
O
O
O

§
Q.
CA

O
i-n
                                                                                                        ID
                                                                                                         CD
                                                                                                         CO
       cn
                                                                                                    3 -*»
                                                                                                   ^ O  r+
                                                                                                   •^j< -j> -j.
                                                                                                   o n>  o
                                                                                                   333
                                                                                                   tO O  (/>
O 0> T3
tt>  3 
C+ Q.
O)    OO
   r- m
o  m r*
o  < *
-*  w n>
-5  — '(B
n>  cn 3
o
rf O O
CD  -h O
Q.   3
     So-
     C
o    
   Q) C*
                                                                                                   o o
                                                                                                   o
                                                  3-63

-------
       Figure  3-20
                    78
Relationships Between Idle Power  Combustion  Efficiency

..evels of CO and HC Emissions, and  Engine Cycle Pressure

Ratios for Aircraft Gas Turbines  Burning Kerosene
c

                                                                                      40
                          Engine Cycle Pressure Ratio Rating
                                          3-64

-------
Figure 3-21
             79
      Effect of  Inlet Temperature on CO  and
      HC Emissions (501-K  Industrial Turbine,
      Constant Speed, JP-5 Fuel)
      EMIIIOM
       (PfW
70-1
CO
50
40
50
20
10
 0
                                      COMMUM*
                                      UUT TWMATUM
                     1000
               2020     5000
              SHAFT HOMEPOMEK
                                            HV340CAUONS
  Figure 3-22
                 80
    EMMIOM
     (PPW
n
70-
60-
50-
40
SO-
20-
10-
 0
        Effect  of Inlet  Pressure on  CO and
        HC Emissions (501-K Industrial  Turbines,
        Constant RPM, JP-5 Fuel, 100 F Compressor
        Inlet Temperature}
                 . 13.7 KM

                                     W.7 PSIA
                             2000      3000
                             SHAFT HOMINHM
                           3-65

-------
in decreases in HC and CO emissions over the engine load range.   The effects
of variations in 1nl» c temperature will  also vary depending on engine and
combustor design so the data of Figure 3-21  cannot be extrapolated to other
engine designs although they do indicate the general  trends.
     Figure 3-22 shows CO and HC emissions versus turbine load for the 501K
engine for two difference compressor inlet pressures  - 13.7 psia and 14.7 psia.
This figure indicates that the effect of this small change in ambient pressure
did not significantly change the CO or HC emission levels.
     3.2.3.3.3  Effects of fuels used in gas turbines  -  Another factor which
may affect HC and CO emissions from a given engine is the type of fuel
burned.  These effects can be somewhat alleviated by proper design of the
combustor to burn specific fuels.  Figure 3-23 shows CO emissions versus load
for a 26 MW gas turbine produced by Westinghouse Electric Corporation when
burning a heavy distillate fuel, No. 2 distillate fuel, and natural gas.  The
data do not indicate any significant differences (small differences indicated
could be attributed to data scatter) in the levels of CO emissions when
different fuels are burned although the CO emissions when burning heavy
distillate fuel are consistently above those when burning the other fuels.
Figure 3-24 shows HC versus load for the same turbines and fuels and again no
consistent trend is evident.  Figure 3-25 shows emissions versus load for the
Detroit Diesel Allison 501K engine when burning two different fuels,.  The
data indicate a slightly higher level but again not significant increase of
CO and HC emissions when burning the heavier No. 2 oil and operating at full
load.  At low load, the CO emissions when burning No. 2 oil are significantly
Mgher (about a factor or two) than the emissions when burning JP-5.
                                     3-66

-------
               81
 Figure  3-23      CO Emissions vs.  Turbine  Load  for
                    Various Fuels
           CO
         4O-
         30-
                        D      19

                         LQAD(MW)
                                         •HEAVY DISTILLATE FUEL
                                         •*« DISTILLATE FUEL

                                         A NATURAL QAS FUEL
                  23
                Qp
Figure  3-24   \°*   HCg  Emissions  vs.  Turbine Load for
                     Various  Fuels
        rm  H/C«
         to
         03-
                                  • MWWV 08TILLATE FUEL
                                  • «2 DISTILLATE FUEL H/C«
                                  A NATUNAL OAS FUEL  H/C,
                         e     is
                          LOAD (MM
                                      20
                  DO
    Figure  3-25     Emissions vs.  Load for a Turbine
                      Burning  Two  Different  Fuels
EMIJiIout
  
-------
     Some experiments have been performed when using methanol  as a gas turbine
fuel.  Turbo Power ar.J Marine Systems, Inc., ran tests on an FT4C-1 gas
turbine and the CO versus power output curves for this engine when burning
methanol and No. 2 distillate are shown in Figure 3-26.  Since this engine was
designed to burn distillate fuel, the flow capacity of the fuel system when
burniir; methanol did not allow operation of the turbine at its rated load of
29 MW.  The data indicate that CO emissions are about 40 - 50 ppm greater
when burning methanol and this difference appears relatively constant over the
8 to 20 MW load range.  Data on HC emissions were not reported but since CO
emissions at the higher loads were low, one would expect HC emissions to also
be low since their reaction time is less than that of CO.
     Coal gas is an emerging fuel which has potential use for firing in gas
turbine combustors.  Pillsbury and Lin of Westinghouse Electric Corporation
have obtained emission data from a full scale coal gas combustor test rig
(for use in the Westinghouse W 501D) turbine when burning synthetic coal gas
with various heating values and at a combustor inlet temperature of 450 K
(350°F).    These tests were performed with the combustion rig initially
simulating mid-load operation and then reducing the fuel flow rate to simulate
the idle mode of operation.  Data from these tests are shown in Figures 3-27
and 3-28, which show CO emissions versus combustor exit  temperature  (which is
proportional to load) for synthetic coal gases with heating values varying
from 200 Btu/scf to 90 Btu/scf and for No. 2 distillate  oil.  Figure 3-27
she s that, at mid-load operation, all three gases ranging from 105 Btu/scf
to 200 Btu/scf burned with CO emissions of 50 ppm or less, that emissions
ifcreased with decreasing load and that CO emissions increased with decreased
fuel heating value.  Figure  3-28 shows essentially the  same results for the
                                    3-68

-------
Figure 3-26
           84
CO Emissions vs.  Power Output for a
Gas Turbine Burning  Methanol and No. 2
Distillate Oil
      200
     *
    0160
    Ift
    CO

    a  80
    Q.
    •

    8  40
\
\J
\
\


Methi
\
\
\


not f u

V
•^ 	 1
No.

j|


I dist.




fuel





          8   12    16   20   24    28
                 Power output - MW'S
                          32
                      3-69

-------
 Figure 3-27
               86
   600-
   •soo-
400-
   300-
S
i  200
O
    100
i
i
                 Effect of  Heating  Value on  CO Emissions
                 from Gas Turbine Combustors  Burning  Coal
                 Gases of 105 to 200 Btu/scf
                                  O TESTA,  7 45MJ/SCM.(200BTU/SCF)

                                  X TESTS. 4. 84 MJ/SCM.II30 BTU/SCF)
                                  O TEST C,  3.91 MJ/SCM, (105 BTU/SCF)

                                  ORY  FUEL GAS
     eoo    wo    eco    900    1000    ROOK
           OOMBUSTOR EXIT TEMPERATURE
            •00)
                   (10001
                        (1200)
(1400)
(ISOOJCF)
   Figure 3-28
               87
  t«00
                 Effect  of Heating  Value on  CO Emissions
                 from Gas  Turbine Combustors Burning  Coal
                 Gases of  90 to  105 Btu/scf

a TEST o
A TEST E
0 TEST F
6 TEST G
MJ/SCM
3 9i
3 95
3 as
3 35
BTU/SCF
(105)
( 106)
(93)
(9O)
•/. M20; VOL
0-
18 0
3.0
68
• NO. 2 DISTILLATE OIL
             600    90O     KXX>    IIOO
            COMBUSTOR EXIT TEMPERATURE
                                     1200 K
     (000)    (1000)    (I20O)    (MOO)   (1600) CF)


                          3-70

-------
105 and 106 3tu/scf gases as that shown for Test C 1n Figure 3-27.  Tests



F and G as shown In Figure 3-28 show a rapid Increase 1n CO emissions as



gas heating value drops.  Additional test data (F & G) were not available



because of combustor flameout at lower firing temperatures.  These tests



also showed that the burner combustion stability was poor after the point



where the CO emissions for each test started to rise rapidly.  For comparison,



the CO emissions from this combustor when fired with No. 2 fuel oil are also



shown on Figure 3-28.



3.2.3.4  Factors Which Affect Emissions of Nitrogen Oxides (NO )  -   Nitrogen
              "" "LIL^^  - - ,--  -    .   ._mi-._Mmi_«r      i— n  _   _.__..    ._         ^


oxides produced by combustion of fuels in stationary gas turbines are formed by the



combination of nitrogen and" oxygen in the combustion air ("thermal" NOx) and


from the  combination  of  niltrbg"eh~Tn~ThVTueT with  oxygen  from the combustion


 air ("organic  NO ).
                 J\



     3.2.3.4.1  Thermal NO  and the variables effecting its formation - The
                          A


NO  formation mechanism by the combination of atmospheric nitrogen and oxygen
  J\


(thermal NO ) for fuel lean combustion in the combustor primary zone is
           /\


commonly described and modeled using the Zeldovich mechanism which is described

                             OQ QQ Oft Q1
by the following equations:  00«ey'yu»yi


          a.  02   5=£    20


          b.  N2   ^r    2N



          c.  N + 0  ^  NO



          d.  N + Op -^ NO + 0



          e.  0 + N2 -^  NO + N



The reactions were expanded by others to include an intermediate  hydroxide


                                      92 93
molecule (OH) in the fuel rich region:  '



          f.  N + OH  ^ NO + H
                                    3-71

-------
and to include the intermediate product of hydrogen cyanide  (HCN)  by
combination of a nitrogen atom with a hydrocarbon molecule 1n  the  fuel
                     9'' 95
rich combustion zone.   "
        g.  N2 + CH  -5-fr   HCN + N
        h.  N + OH   -^   H + NO
      These reaction mechanisms are commonly used by the manufacturers  of gas
turbines in computer models to predict emissions from their  conventional
combustors which operate at, or near, stoichiometric combustion conditions.   '
They are not valid for combustors which may be developed to  operate at  off-
stoichiometric conditions (such as fuel rich combustion) and thereby result
in reduced NO  emissions.
             A
      The Zeldovich equation for the formation rate of thermal nitric oxide
               98
(ppm/m sec) is:
                     -  kP°'5  X.  °'5  XM   T  -1  e-122>000/Tf
                                   1-    X
                                                                         'NO
*02       N2   'f
         where:
         K = constant of formation
         P = combustor pressure
         Tf = flame temperature
         X = the volumetric fraction of the substance (i.e., 02; Np^; NO)
        (XNQ ) = nitric oxide equilibrium concentration
            e
      Equation (i) shows that the thermal nitric oxide formation rate is
extremely sensitive to the flame temperature -- increasing exponentially
wi«n increases in flame temperature.  From this equation, we can also observe
that the formation of NO varies with the square root function of the combustion
pressure and with the Np, Op, and NO concentrations.  As the concentrations of
nitric oxide reaches its equilibrium value, the bracketed figure (equation i)
becomes zero and thus the nitric oxide formation rate becomes zero.
                                    3-72

-------
     The thermochemlcal equilibrium quantities of NO that can be generated


depend strongly on the combustion temperature levels and on the availability


of oxygen.  Thus, the equilibrium quantities of NO produced Increase rapidly


with Increases 1n combustor Inlet temperature and as the fuel to air


equivalence ratio approaches values of 0.8 to 1.0.  Figure 3-29 illustrates



the thermochemical equilibrium quantities of NO  which can be produced in
                                               /\

combustion processes.  Fortunately, the quantities of NO  emissions generated
                                                        /\


by gas turbines are limited by the short residence times of the hot combustion



gases within the engine combustors and by the rapid quenching of these gases


with the dilution air.  Therefore, very high combustion efficiencies can be



attained without generating the thermochemical equilibrium quantities of


NO.100' 101> 102> 103  Some typical nitric oxide formation rate data for gas


turbines are shown in Figure 3-30.  Figures 3-30 and 3-31 demonstrate the


tremendous effect of combustor inlet temperature on the formation rate of NO.



Figure 3-30 also shows that NO formation rates Increase as combustor inlet


pressures increase and that operation of the combustion process at equivalence


ratios less than or greater than stoichlometrlc (equivalence ratio = 1.0) will


result in reduced formation of NO.


     Humidity contained 1n the combustion air will effect NO  emissions by
                                                            A

reducing the combustion flame temperatures.  Figure 3-32 illustrates the


effect of humidity and ambient temperatures on NO  emissions from one gas
                                                 A

turbine.  NO  emissions are shown to increase with increasing ambient temperatures
            y\


(0 percent relative humidity curve) and to decrease when humid air is used


(100 percent relative humidity curve compared to the 0 percent relative


humidity curve).
                                       3-73

-------
  in
  C
  O
  O)
  E
 jQ
  cr
  ro
  CJ
  o

  S-.
  QJ
en
en
 oo

 QJ
 S-
 3
 CD
                                                                                  (U
                                                                                  o

                                                                                  01
                                                                                  cr
                                                                                  w

                                                                                  4)

                                                                                  O
                                                                                  N
         O
         O
         PO
o
m
esi
O
O
O4
o
o
o
in
                     go spunoj  QOOI
                                                                      ON  30
                                                3-74

-------
      Figure 3-30'1 °3  Formation Rate Data for NO
10.01
v» MO
  100
   so'


   10
    s
                                 UitUl Mixture Condition*
                                         1000* I. M Atw>«.
                                         tM* K. SO At
                                         TSO* X. 10 At

                                             K. 30
    •.S    0.4
                              i.o   i.a    1.4
                                                       l.ft
           FUa-AIR EQUIVALENCE  RATIO
                         3-75

-------
Figure 3-31     NQ   emissions vs. Combustor Inlet Temperature as
                Presented by Lipfert  (corrected to ambient temperature
                = 80UF and humidity =  0.01  gm H20/gm  dry air)
                  .100       600       goo


                  CONIUSTOI INLET T[Nf[I»Tim (V)
                           3-76

-------
Flaure 3-32     Tne Effect of  Ambient Temperature
  y               and Humidity on NO   Emissions  from
                  a  Turbine
                                 1001
                             «CL»tIVI
                   «*no • u

            COMMISSO c'Mcimci • .11
        0     20     40     «0     10     tOO     lit



                  • KIKNI IIKKIATUIH (-F)
   Figure  3-33  °6   Influence of  Residence Time on
                      NO^ Emissions
          NORMALIZED TO

  COM8USTOR WLETPRESSURE 60PSIG(4.0ATM)

  COM8USTOR INLET TEMPERATURE 600*F(3I5*C)

  TCOB.OUT-TCOM8.IN«AT NOM»I050*F(570*C)
    NOX

    PPMIVO.)
                      -ANCHOR POINT
uu

7O
60









































































^









V








if









•^








*







/

•^.





^












^.


•
'





^


20 — ^2§ 30 35
                                       CURVE
                                     CALCULATED
                                     FROMANCH.
                                       POINT
                                      KG/SEC
                                     AH FLOW.
                        3-77

-------
     Increased residence time of the gases 1n the combustor will  result 1n



Increased emissions o* NO  until the residence time becomes so great that
                         /\


the NO equilibrium value 1s reached (see equation 1  -  bracketed quantity).



Data on the exact effect of residence time on NO  emissions from a given
                                                /\


gas turbine combustor are scarce primarily because other variables (such as



mass flowrates) change with a change in residence time.  The data in Figure 3-33



show NO  emissions versus air flow and indicate that increased residence time
       J\


(which is inversely proportional to air flow) results in increased NO  emissions.
                                                                     A


     The specific design of the combustor such as fuel and air injection



methods and combustor cooling air flows will also effect emissions.  Discussion



of some of these variations and their effects on NO  emissions can be found
                                                   x


in Chapter 4 of this document.



     There have been many attempts to develop correlations between NO  emissions
                                                                     A


and the many variables (such as temperature, pressure, humidity, residence



time, air to fuel ratio, etc.) which effect these emissions.  Sullivan and



Mas prepared a paper in July 1975 titled, "A Critical Review of NO  Correlations
                                                                  /\


for Gas Turbine Combustors" in which they reviewed the semi-empirical and



empirical correlations which had been reported in the literature.     They



concluded that an adequate data base does not exist to evaluate the various



NO  correlations.  They also concluded that:
  J\


     a.  The effect of inlet temperature, pressure, fuel to air ratio and



mass flow on NO  production differ significantly between the various
               /\


correlations.



     b.  The independence of the pressure, inlet temperature, fuel to air



ratio and airflow effect on NO  has not been demonstrated.
                              /\
                                     3-78

-------
     During the process of developing standards of performance for
stationary gas turbines, several correlations were provided to EPA and these
are summarized 1n Table 3-9.  Equations 1 and 2 1n Table 3-9 were submitted
with the recommendation that they be used to correct emissions from gas
turbines, regardless of manufacturer, to a common set of reference conditions.
Manufacturers of gas turbines were quick to respond that these equations were
not applicable to their engines and stated that, in their opinion, individual
equations would probably have to be developed for each gas turbine and
combustor combination.  Equations 3, 4 and 5 of Table 3-9 were submitted with
additional data to show their applicability and accuracy when applied to
engines of the specific manufacturer.  Data presented by the manufacturers
indicate that equation 3 can correct emissions to an accuracy of ^10 percent,
equation 4 has a reported accuracy of j^7.5 percent, equation 5 and equations
developed by Detroit Diesel Allison Division of General Motors also have been
shown to predict emissions very accurately for specific machines.
     Despite their obvious differences, the equations reported in the
literature and those shown in Table 3-9 do have some common characteristics.
These common characteristics are summarized as follows:
     a.  Most pressure dependence data supports the conclusion that the
correction factor for the effect of pressure on NO  emissions is of the form:    '
MH   -  wn      / po reference\
N0x  "  N0x OBS I  J   •     ^  I
  x       x uoi ^ p  observea J
                                          0.5
                             3
          Where P- equals the compressor discharge or regenerator discharge
          pressure.
     b.  Most humidity dependence data supports the conclusion that the
correction factor for the effect of humidity on NO  emissions is of the form:
                                       3-79

-------
                                                      Table 3-9. NOX  CorrtltMoM  ftr tat Turblrwi
      Source of Correlation
                                                                                                                      Oiflftltlen af T«n»i
1.  Contention and  Finis CoMUtM,
    Annrican Socltty of Mechanical
    Engineers''™
      N0
                                                     3obs
                                          C,D(f/a ref - f/a obs) ,   (T3obj)tf/a obs) - T3r(jf(f/a refj
                                                                  * Coipraisor dlichtryt or rtgcntrator d1senary*
                                                                    prttiur*
                                                                   specific humidity
                                                               T. - M»pr«iior dltcnargt or rtgtntritor
                                                                    temperature
                                                               f/a - fu*l/«1r ratio by wtlght
                                                               3{ • oxygtn concentration fay volume
                                                               C, D, E • empirical constants
                                                               obi • observed
                                                               ISO • International Standard Organization, standard
                                                                     ataoiphtric conditions: 15'C (519%),
                                                                     1.013 Bar. (14.7ps1a), 60S relative humidity
                                                                     (.00633 Ib H20 )
                                                                                                                    Ib air
                                                                                                      ref • reference
                                                                                                         CQ * determined fro* connnly used gas  turbine
                                                                                                              performance calculations or curves.	
    American National Standard
    Institute109
        NO,
     ,.JO )    [  10 ]°'5  O-005^15 - Tot*)   19(Hobs -.0064)


[20-9-15     ]  - 18640 [&•&*    ] , „
L20.9-(02)obs          1 + (f/a)obsJ
                                                                P • burner Inlet pressure 1n atmosphere
                                                                T • burner Inlet temperature In *C
                                                                H • specific hu»1d1tv~gm rUO/gm air
                                                                0, " exhaust oxygen, percent by volume
                                                                f7a • fuel/air weight ratio
                                                                N - fuel bound nltrogen'M'ercent by weight
                                                                obs - observed during tests
                                                                Base conditions for equation
                                                                                                          10 atmosphere (I.e.  turbine has  10/1  compression
                                                                                                                         ratio)
                                                                                                          315*C (approximate heat of compression  for
                                                                                                                 10 atm.)
                                                                                                      H - .00633
                                                                                                      SO, * 15 S base as suggested by EPA  in previous
                                                                                                            draft documents            b
                                                                                                      tN - NO, fuel has essentially no found N
    General  Electric110-
    "A Constant Residence  time
    equation for Correcting NO
    Emissions to a  Set  of  Gas
    Turbine  Operating Conditions-
                                    3ref
                                                                NO „« • N0_ measured at reference conditions of
                                                                  xref   f/l. P,. H. T3.
                                                                    (N0xref •acl"ne x) - 1 for single machine model

                                                                     NOxref machine Y
                                                                P, •  compressor or regenerator discharge pressure

                                                                T, •  Compressor or regenerator discharge temperature
                                                                H  - specific hum1d1ty~*t H,0/wt dry air
                                                                Ref « reference condition
                                                                f/a • fuel/air ratio by weight
                                                                1H -  H - Href
                                                                            ref
                                                                            f/a ref
                                                                              3ref
                                                                C,  •  .00214 • empirical constant
                                                                Ci •  39,320 • empirical constant
                                                                applies to constant speed simple cycle and regenera-
                                                                tive machines and simple cycle variable speed
                                                                machines.             	  	
4.  Pratt & Whitney -  Turbo
    Power and Marine!'1
          'NOx
         i        r(Pp) ref-,0.5   0.003l3[(Tb1)ref-(Tb1)obs)]
         'NOxobs  l(Pb) obsj    e
                                                                                                       El • emission Index »-—• g NO, emitted
                                                 (e!9(H - .00634)
                                                                                                 Kgfuel
                                                                          Pb * burner pressure
                                                                          Tbi »  temperature at burner Inlet
                                                                          H • specific humidity
                                                                          obs •  observed conditions
                                                                          ref *  reference conditions	
b.  Mestinghouse
                112
         N0x
         WJref
                                                      ,0.5  [(T-Trtf t (tT-ATref)  +  12(H  ref-H)]
                                                                 P • compressor discharge  pressure  -  Psla
                                                                 T • Compressor discharge  temperature -  °f
                                                                 tT • combustor temperature  rise  -  °F
                                                                 H • specific  humidity
                                                                 ref m values  of parameters  at  the  reference  condition
                                                                 NO  - ppm(v)  - expressed  as NO,
6.  Detroit Diesel  Allison
    Division of General  Motors
113
The humidity correction  factor used 1s  19(H-.00633)
Other corrections  were obtained by computer analysis of
test data to provide  equations for correlating emissions
from the 501 K engine.	
7  Solar
        TT4~
                       TTT"
         The equation  which GE proposed for correcting NO  emissions
         (see Item 3 of this  table) 1s not applicable to Solar engines.
         Solar believes the constants in that equation will differ
         "depending on whether the engine is single shaft, two shaft,
         simple cycle, or regenerative cycle as well as combustor type.
8.  Garrett Alreasearch
                                       Emission data not sufficient to indicate a preferable
                                       correlation.
                                        Use humidity correction:           nou    nnei/i\
                                          NOX corrected = (NO  measured) euaM " •uubj<1'
                                        Temperature correction?
                                             Upfert curve (from literature) fits their engines  fair
                                             well.     	
 9.   Turbod"nc
              116
         Humidity        NOX - (NOX measured)  e15'73 'H'  •00t>J4)

         Temperature:  "It appears that for each gas turbine  there
               exists a unique relationship between ambient  temperature
               and NO  emissions."  They suggest that each manufacturer
               be permitted to establish the NO  -  temperature  curve fc
        	each specific engine - combustor combination.	
                                                                          3-80

-------
          Where K, a dimension!ess constant, has been shown to be equal


                    119
          to 22 + 8.     Review of the literature and the tabulation 1n



          Table 3-9 reveals that K=19 1s the value most commonly chosen



          although some manufacturers (as shown in Table 3-9) have developed


                                                120  121
          a specific constant for their machine.   '      Ideally, the



          development of a specific constant for each machine is the better



          approach since it has been analytically calculated that the correction



          factor for humidity will vary with compression ratio, which is a



          function of machine design, as demonstrated in Figure 3-34.



     c.  Sullivan and Mas plotted NO  emissions versus ambient temperature
                                    ^


at constant combustor outlet temperature for the various correlations found



in the literature.  These data are shown in Figure 3-35 which shows that all



the correlations except one predicted a small.variation in NO  due to ambient
                                                             /\


temperature changes.  This relative insensitivlty of NO  emissions to ambient
                                                       ^

                                                                        124
temperature variations can be explained by several compensating effects:



          (1)  As the ambient temperature increases, combustor pressure



          and air to fuel ratio decrease, thus tending to decrease NO  emissions.
                                                                     ^


          (2)  Concurrently with (1) above, the combustor inlet temperature



          increases and the air flowrate decreases thus tending to increase



          NO  emissions.
            /\


Based on the preceding, one can conclude that the important ambient conditions



which effect NO  emissions from stationary gas turbines are the ambient pressure
               X


and humidity with ambient temperature having relatively little effect.  This



conclusion  is not, however, substantiated by the data developed by Detroit



Diesel Allison Division of General Motors Corporation as shown in Figure 3-36.
                                       3-81

-------
     Figure  3-34122   Correction  Factors for  Ambient
                        Humidity
T)


I


-------
Figure  3-35
             123
       NO   Emissions vs.  Ambient Temperature
       as  Predicted by  Various  Correlations
     400
      300
    2
    O.
    0-
     ,200
    c
    z
      100
           m*iso • 2.5
           T4 -1235 K (1800'F)
           0% RELATIVE HUMIDITY
                    SIMPLE CYCLE
                    CONSTANT SPEED
	 MARCHIONNA, DIEHL. TROUT
	DAVIS, MURAD. WILHELM
	 VERMES
	 MARSHALL
	COHEN
          260  270   280
                K
               290
                                300
310   320
                                      _L
               i   i   i  T—r   i   i   '   i
              20    40    60    80    100
                  TEMPERATURE (AMBIENT)
                            •r
                                 120
                            3-83

-------
               125
   Figure  3-36      NO  vs. Shaft Horsepower for Various Compressor
                     Inlet Temperatures  (501-K Industrial Turbine,
                     constant speed, JP-5  Fuel)
             150-
       des   100
of Nitrogen
  — ppm
             50-
                            CWMIUO* INLET
               0     1000    20CO     3000     1000
                           Stun HowtwM
                               3-84

-------
This figure shows that NO  emissions about double when the compressor inlet
                         X


temperature is varied from -20°F to 100°F changing from approximately 70 ppm



to 150 ppm at an output of 4000 horsepower.  Based on this, the different



corrections for temperature reported in the literature and Table 3-9 and



comments of manufacturers it is concluded that the specific design of the gas



turbine will affect the nature of the NO  ambient temperature relationship
                                        /\

                                                                        126 127 128 12'
and, therefore, a unique relationship will exist for each turbine model.   '   '   '



A correction factor for the effects of ambient temperature on NO  emissions which
                                                                A


is universally applicable to all gas turbines does, therefore, not exist.



     Also, it must be remembered that the correlations shown in Tabl« 3-9



and commonly reported in the literature were developed for combustors which



operate at stoichiometric conditions in the primary zone and, therefore,



should not be applied without verification to low emissions combustors which



commonly utilize some means of suppressing flame temperature below stoichiometric



values.  For example, NO  emissions from combustors which operate with a fuel
                        /\


rich primary zone have been found to decrease with Increases in fuel to air



ratio while NO  emissions from the more conventional combustors with lean
              /\


primary zones have been found to increase with increased fuel to air ratio.   '



The slope of this increase for the conventional combustor depends on specific



combustor design features and thus generalized fuel-air ratio correlation


                                132
factors have not been developed.     For this same reason, when using correction



factors for extrapolating NO  emissions data acquired at one set of ambient
                            J\


conditions to any other ambient conditions such as ISO conditions, one should



require that the turbine be operated at a reference air to fuel ratio or firing



temperature (such as full load turbine design conditions) while emissions



measurements are being made.
                                      3-85

-------
     3.2.3.4.2  Organic NO  and variables affecting Its formation  -  Organic
                          A

NO  is formed during combustion by the chemical  combination of the nitrogen
  A
                                                              1 O VI  O1CT  1 O £  1 *n

atoms contained in i,,ie fuel molecule with oxygen from the air.        '


Although many papers have been written on the conversion of fuel bound nitrogen


(organic nitrogen to NO ), the exact mechanism of NO  formation is not known   »I39»
                       A                            X

Generally, organic NO  is only a problem when burning residual oils, blends,
                     A


some crude oils, or heavy distillate fuels which have high nitrogen contents.   '


Most light distillate fuels have less than 0.015 percent by weight nitrogen,


crude oils generally have less than 0.2 percent and residual oils can go as

                  14T  144
high as 2 percent.IHO' IHH


     The maximum organic NOX that will be produced when burning a given fuel


in a gas turbine, assuming all fuel bound nitrogen is converted, may be

                                       145
calculated from the following equation:




                                 Nf    X    Wf


          NO  (ppm by volume)  =  	     X 30   X  28.95   X  106,


                                 Wf + Ww + Wa       "*       3°



          where Nf  =  fuel-bound nitrogen, percent by weight (expressed as  a


                       decimal


                Wf  =  fuel flow rate, Ibs/sec


                W   =  combustor air  flow rate, Ibs/sec
                 a

                W   =  water  injection flow rate, Ibs/sec
                 W

                30  =  molecular weight  of NO


              28.95  =  average molecular weight of combustor  exhaust  gas


                14  =  atomic weight  of  nitrogen


          This simplifies  to:
                                  3-86

-------
                 N- X W,/W  X 2.0679 X 106
                  T    TO

          NO  =  	        (ppm by volume)

                    (i + wf/wa + ww/wa)




     In their experiments with a combustor from the MS 5000 series gas



turbine, General Electric Corporation found that the actual conversion of fuel



bound nitrogen varied with the nitrogen content of the fuel and to a lesser



extent with the fuel to air ratio at which the gas turbine was operated.



They therefore performed experiments using the MS 5000 series combustor



operating at a constant fuel to air ratio of 0.021 to develop the curve shown



in Figure 3-37.  This curve shows that the percentage of fuel bound nitrogen



converted to NO  decreases as the fuel nitrogen content increases.  Using
               A


this curve and the equation given above for calculating the NO  contribution
                                                              A


assuming 100 percent conversion of the fuel nitrogen, one can then calculate



the NO  contribution of the fuel to the total NO  emissions from the MS 5000
      X                                         A


gas turbine operating at a 0.021 fuel to air ratio.



     The general applicability of the curve shown in Figure 3-37 to combustors



with different designs has not been shown.  For example, data developed by



Dilmore of Westinghouse shows that for a fuel with 0.79 percent bound nitrogen



the yield factor was about 20 to 30 percent compared to the 40 to 50 percent



shown for the G.E. machine in Figure 3-37 (both machines were operated at



firing temperatures in the 1800°F range).     The data in Dilmore's report



(Reference 147) also show that the percent conversion of fuel bound nitrogen



decreases with increased firing temperature while G.E. data indicates that the


                                                                148
percent conversion increases with increasing firing temperature.     It has



also been reported that the yield of NO  from bound nitrogen in the fuel is


                                                                            149 150
less for fuel rich combustion than for combustion with leaner fuel mixtures.   '



As pointed out earlier in this section, the exact mechanism of NO  formation from
                                                                 A


fuel bound nitrogen is not known.  The variables which affect the emission of



organic NO  are not defined and further research is needed in this area.
          A




                                     3-87

-------
   1.0

z
LU
CD
O
K
5  0.8

a
            YIELD CALCULATION IN
            THIS AREA REQUIRES
          MEASUREMENT ACCURACIES
          IN EXCESS OF INSTRUMENT
                RESOLUTION
a
CO
O
tr
o
K
u.
a
_i
LLJ
 X
o
   0.6
   0.4
          YIELD FRACTION =
                           MEASURED ORGANIC NOX
                                                                          VARIATION DUE TO
                                                                      MEASUREMENT TOLERANCES
   0.2
                           CALCULATED ORGANIC NOX
                           FOR COMPLETE CONVERSION
                              LABORATORY TEST DATA AT FUEL/AIR RATIO = .021
                                          DRY COMBUSTION
    OL-
    0.01
                 0.02
0.04     0.06   0.08  0.1            0.2

  FUEL-BOUND NITROGEN, percent BY WEIGHT
0.4
0.6    0.8  1.0
                Figure 3-37^ 46  I\IOX yield fraction vs. fuel-bound nitrogen content.
                                              3-88

-------
     3.2.3.4.3  The effects on NO  emissions of burning various fuels 1n



statjonary gas turblnes  -  The effect of fuel bound nitrogen on NO



emissions from gas turbines has been discussed in section 3.2.3.4.2 of



this report and, therefore, will be not discussed again.



     It has been known for some time that burning natural gas instead of



number 2 distillate in a given gas turbine can result in reducing NO
                                                                    A


emissions by as much as 50 percent.     This has been theorized to occur



because the combustion of liquid fuel droplets proceeds in the reaction



zone at fuel to air ratios which result in localized peak flame temperatures



(which are conducive to greater NO  formation) while the combustion of
                                  ^


natural gas occurs at lower, more uniform temperatures.   '     The relative



levels of NO  emissions from one gas turbine firing natural gas, heavy
            ^


distillate (-<0.05 percent nitrogen) and number 2 distillate (-<0.05 percent



nitrogen) are shown in Figure 3-38.  No significant difference in NO  emissions
                                                                    y\


for the heavy or number 2 distillate fuels was observed.



     Figure 3-39 shows NO  emissions versus firing temperature for a simple
                         /\


cycle gas turbine produced by General Electric and burning crude and distillate



oils.  Again, no significant difference in NO  emissions is observed.  No
                                             rt


other data for NO  emissions from gas turbines burning crude oils are available.
                 A


     Estimated emissions from a FT4C-1D gas turbine manufactured by Turbo



Power and Marine Systems when burning number 2 distillate and methanol are shown



in Figure 3-40.  The NO  emissions are reduced by about 75 percent when methanol
                       y\


fuel is burned rather than the number 2 distillate.  Data supplied by General



Electric Corporation from tests performed on a production combustor for their



MS 5000P gas turbine show a similar trend, indicating a 63 percent reduction



in emissions of NO  when burning methanol as compared to number 2 distillate.
                                        3-89

-------
Figure  3-38
            154
NO   Emissions  vs. Load for a
Turbine Burning Various Fuels
        MOi
        ISO
        oo-
        so-
                                     • WOT OSmJLATI FUEL
                                     • +t WSTILLATC FUEL
                                     ANATIMAL OAS FUEL
                      10     IS

                       U0«0 (M W)
                            3-90

-------
       OJ
      T3
       3

      O

       I

      CM
       4J
       O
       O

       CM
  TJ
  s
  O
   I

  CM
 0)
4->i
 lt3


'•r-
•«->
 in
•r-
Q

 I

CM
     O
     0)
    a
    «fr
x  O
LT>
If)
 co
  i
 co

  o>

  3
         0)


         4->
         (O
         J- C31
         (H C
         O-'i-
         •-3^
          01   tt>
          C  OJ 3
         •^-  C U-
          t. -r-
         •i- JD  O (/>
C 1^ -r-
O l/> O

 UJ C
•i- CD «

LU «J 0>
     •O
  XJ- 3
O O 1-
Z <4- O
                          I      I      I—I      I      I      I
                                                                           CO
                                                                           l
                                                                           o
                                                                           o
                                                                           IT)
                                                                           O
                                                                           O
                                                                                0)

                                                                                3
                                                                                •)->
                                                                                It)
                                                                           O
                                                                           o
                                                                           CM
                                                                                 L.
                                                                           O   t-
                                                                           c   u.
                                                                       -hi
                                                                       t§
                  s
                                    o    o
                                    ^J-    CM
                                              O
                                              CO
                                               O
                                               IO
o
CM
                                                                                       O
                                                                                       O
                                                                                       03
                                                           iudd
                               3-91

-------
           156
Figure 3-40     NO  Emission vs. Power Output for

                a Turbine Burning No. 2 Oil  and Methanol
        Tamb - 73°F - 54% Relative humidity
2UU
CM
0 160
u>
I 120
CO
8: so
i
K
i 40
n

0-*"**




"75=—






.AA,,,-
/f' 1





o*=£





^^-Ivid




__ — •
thano






fuel-
      8   12    16   20   24   28   32

             Power output • MW's
                      3-92

-------
     As noted in section 3.2.3.3.3, coal gas 1s an emerging fuel which



has potential for use in gas turbine combustors.  Figure 3-41 shows measured



and predicted emissions of NO  when burning natural gas, number 2 distillate
                             A


and low Btu gases in a coal gas combustor.  The agreement between predicted



and measured values is close for the number 2 distillate oil and the 7.45 MJ/scm



(200 Btu/scf) coal gas.  The predicted NO  emissions from burning the 4.84
                                         A


MJ/scm (130 Btu/scf) coal gas are 3 ppm but this was not verified because


                                                                 159
measurement instruments could not resolve emissions below 10 ppm.     The



curve showing the 11.18 MJ/scm (300 Btu/scf) gas is based on an oxygen blowing



gasifier and, as can be seen from Figure 3-41, the NO  emissions are predicted
                                                     A


to be higher than when number 2 oil is burned.  This predicted high level of



emissions for the 11.18 MJ/scm (300 Btu/scf) gas is caused because the oxygen



blown gasifier results in coal gases with high hydrogen and carbon monoxide



contents which burn at higher flame temperatures than natural gas.     The



reference therefore predicts that high levels of NO  will result if pure
                                                   A


oxygen is used as the oxidizing medium in a coal gasifier, but not if the medium



is air.     Generally, the lower flame temperatures of low Btu fuels (e.g., air



blown coal gas) will result in substantially reduced NO  emissions as compared
                               1 fi?
to natural gas or number 2 oil.          This is also shown in Figure 3-42



for NO  emissions from full and small scale combustors burning natural gas
      A


and coal gas.  Lower NO  emissions from low Btu gas as compared to number 2
                       A


distillate and natural gas have also been predicted based on laboratory



combustor rig tests of a Texaco gasifier (100 - 150 Btu/scf gas) coupled with



a Turbo Power and Marine FT4 combustor rig and the estimated emissions of



NO  for an FT4C-1 gas turbine based on results of these tests are shown 1n
  A


Figure 3-43.
                                     3-93

-------
Figure  3-41
             158
Measured and  Predicted  NO  Emissions  from Burning
Various Fuels in a Gas  Turbine Coal Gas Combustor
                     WOO   UOO   1400   ICOO   IIOO

                  COMBUSTOR EXIT TEMPERATURE. *F

               pool  «MOI  (toot dooa  (ilooi 111001  «joa  IMOOKI
                                3-94

-------
Figure 3-42
i164  NO   Emissions from  Full  and Small Scale
     Cotibustors vs. Combustor exit temperature
     for Gas Turbine  Combustors Burning Natural
     Gas and Coal Gas
        OT
          O.
          IB
        IS-
        0.
        a.
           900
        NATJSAS
        COAL GAS
                          KTG.VALUE
                          M4tn3 (AT S.TR)
                          (BTU/SCF)
                        IBUSTORS:
4.6
(124)
43
(129)
5.0
035)
D.I
(270)
38
980
—
	
A
•
•
—
—
A
D
—
o
X
—
—
—
      1000
1100   1200   1300 .  1400 K
           (1200)    (1400)  (1600)   (1800)

             COMBUSTOR EXIT  TEMP.
                             (2000)'(»F)
                           3-95

-------
     As discussed 1n section 3.2.3.4, there are many design factors which



affect the NO  emissions from a gas turbine.  These factors will also affect
             x


the NO  emissions f urn a gas turbine burning low Btu gas.  For example, the
      A


temperature of the coal gas used by the gas turbine will affect NO  emissions
                                                                  A


by causing them to increase with increasing coal gas temperatures and



any nitrogen compounds within the coal gas may be converted to NO  as discussed
                                                                 A


in section 3.2.3.4.2.  Therefore, the NO  emissions from the gas turbine may
                                        A


be affected by the design characteristics of the low Btu gas generating system.



     3.2.3.4.4  The effect of jncreases in gas turbine thermal eff 1i ciency on



NO  emissions  -  Manufacturers are continuously striving to increase the
~~"~"X


thermal efficiency of their gas turbines.  Figures 3-44, 3-45, 3-46, 3-47, and



3-48 depict the increase in efficiency which gas turbines have experienced



since 1960 and project the efficiencies for the early 1980's (heat rate as used



in the figures is inversely proportional to efficiency).  Figure 3-44 shows



that the efficiency of gas turbines increased by about 40 percent from 1960



to 1975 and projects further increases in efficiency through 1980.  Figures



3-46 and 3-47 illustrate the trend towards higher firing temperatures and



compression ratios to obtain increased efficiencies from simple cycle gas



turbines.  Figure 3-48 illustrates the same trend toward higher compression



ratios, firing temperatures and efficiency for unfired combined cycle gas



turbines.



     Figure 3-49 shows thermal efficiency versus specific output in Btu's



of net output per pound of gas flow for simple cycle, regenerative cycle,



ano combined cycle gas turbines for firing temperatures of 1650°F and 2000°F



and for different compressor pressure ratios.  This figure shows that the



efficiency of simple cycle machines and combined cycle machines increases



with increases in pressure ratios and firing temperatures.  The figure also
                                       3-96

-------
Figure 3-43
            165
        Estimated NO  Emissions  vs. Power Ootput For
        an FT 4C-1 Gas Turbine Burning Low Btu Gas,
        Natural  Gas, and No.  2 Distillate
240

200

160

120

 80

 40

  0
                                         IOW BTU 6*5
           10      15       20       25

                    POWER LEVEL-MEGAWATTS
                                    30
                           3-97

-------
Figure 3-44
           166
            Improvement  1n  the  Heat Rate of Simple

            Cycle  Gas  Turbines  Since  1960
   -16,000 Bta
   -14,000
   -12,000 Btu
•10,000  Btu  S.  .


            m
            


  8.0QQ  Btu  *%  3
            o+f
            _j co
                10

                0)
                                      Little or No

                                      Development
                     Year-


                      1970
                                     Continued

                                     Development
                                             1980
                                               i
                       3-98

-------
   TOO
72
55
                                Efficiency  %
49  46 44 41    38  34
29











100%
U^ Eff.
1

Vh'th Process








Steam _







»•













•












Total ,
Energy













f









1980
Combi n
Cycle
Goal
X

^r



ed




^ With
Boi




ler






















tt











>-

i —





19
1975

Combined Cycl
-
50 N
**•
_100 !



IW
IN




>
Vk

1975 Large
• Utility
Stean
75
Diesel
10 MW


i Tu
^
1


'bin





a

1980 S



mple
" Cycle Goal




1975 Stean
•*^i Turbine
J 50 MW

1975
Simp
Cycl<


e "^
a 6.T
3000    4,000   5,000    6,000  7,000  8,000   9,000   10,000   11,000   12,000

                          Heat Rate  Bta»/kwh
             Figure 3-45
                        167
              Simple Cycle and Combined Cycle Gas Turbine
              Heat Rates and Efficiencies for 1975 and Projections
              for 1980.
                                       3-99

-------
Figure 3-^
        68
            Projected HHV Heat Rates for Simple Cycle
            Gar Turbines vs. Firing Temperature for Various
            Co  resslon Ratios
   I2.nna
   ',1,000
   10,000
  .
(tt
    9,000
              'of Becent
              Technology
              Evolution \
                Target
               Range of
               Development
               Required
Heat Rate
Line
                                      ^Approximate
                                      Level of
                                      Technology
                                            1972
                                    Temp.
                                    Design
                                  I.S.o7~Cond1t1ons
                                  No. 2 Fuel Oil
      1200  1400 1600 1800 2000 2200 2400

          Firing Temperature - °F
                     3-100

-------
            Figure 3-47169  Heat Rate (HHV)  vs. Time of Order for
                          Simple Cycle Gas Turbines

p
b
  14,000

  13^000
r '2,000
u
£ iipoo
or
^ 10,000
UJ
x 9,OOO

  8.0OO
m
                       o
                             A
                               A
— TREND FOR     l-Q0r
   THE BEST HEAT    ^v^--
_ RATE AVAILABLE
   CLARGE FRAME
   GAS TURBINES)
 !•
    I.S.O CONDITIONS
_  NO. 8 FUEL Oil
    BASE LOAD RATING
    I   I    I    I    I
                          ©
                          ©
(£) 6E FRAME 5
0 GE FRAME 7
S ® l91
® ® 251
GO (g) 501
A TURODYNE TYPE n
O AIRCRAFT DERIVATIVE
                                                       : I  CR & 2000'F
                                EXTRAPOLATION
                                 OF RECENT TRENDS
                                                    ^O'.l  CR 8 2500*F
                                       I    I    I    I   I
            1968    1970    1972    1974    1976    1978    1980    1982
                               3-101

-------
         Figure 3-48
                    170
                             HHV Heat Rate vs. Time of Order for
                             Unfired Combined Cycle Generating Units
  13,000
^12,000
 « 11,000
 0010,000
 x
 a
 h- B,000L
 ee
 H 7,000
 tu
 r 6,OOO
                      I. GE STAG 100
                      2s GF. STAG 200
                      3 GE STAG 300
                      4 GE. 3TAG 400

          Best  available coal  fuel
G *' PACE
A TURBODYNE
O AIRCRAFT DERIVATIVE
   I.S.O.  Conditions
   No. 2 Fuel Oil
          LJ\«iiJt«l4Fl-tllV4ulvwSSUi  JW^.1       _       JT\J-J^J
          Central  Station heat rate",'   Base Load "atirtg
                         -*«,-	^
                     Extrapolation
                   of Recent Trends
                                            20'I  CR 8 2500*F
5'000668   WTO    072
                             !.  L._i	1.
                             1974    1976
    .__
   1978    I960   I98X
                            3-102

-------
                 i 70
         Figure 3-49    Gas Turbine Plant Performance for Combined,

                     Regenerative and Simple Cycle Gas Turbines
50r
•p
245
>-40
u
z
UJ
u-
UJ
  25
  20
           FIRING TEMPERATURE

           ---- 2000°F

           - 1650'F



                                       NUMBERS  ON CURVE*

                                        ENOTES PRESSURE
                         SIMPLE CYCLE
       60   80     100   UO    140

            SPECIFIC OUTPUT (BTU/LS)
                                    "160
                         3-103

-------
shows the relationships between thermal efficiency, firing temperature,



and compression rat'o for regenerative machines and indicates that optimum



efficiency for these machines occurs at higher firing temperatures with



moderate compression ratios.  This high efficiency at moderate compression



ratios is due to the fact that the transfer of heat energy from the exhaust



gases to the combustor inlet air (compressor outlet air) depends on the



temperature differential between them and, therefore, a low compressor



temperature rise (which corresponds to a low pressure ratio) provides optimum



efficiency.



     In section 3.2.3.4,1, it has been shown that the production of thermal



NO  increases asympototically with increases in firing temperature and also
  /\


increases as a square root function of the combustor inlet pressure.  Therefore,



the trend towards more efficient gas turbines with higher firing temperatures



and compression ratios will result in much greater production of NO  from a
                                                                   J\


conventional type gas turbine combustor.



     3.2.3.4.5  The effect of gas turbine configuration and operation on NO
                   " " "    i "--- J   -    -  --   •— — ]       - J-IL .   ._    _n. i _i .._L - i— .—LJ— •--           j\


emissions  -  The same basic turbine can be used in either a simple cycle



or combined cycle configuration.  Emissions from identical turbines used



in the simple cycle and combined cycle (unfired boiler) will be identical



at a given turbine operating load.  For identical turbines used in the simple



cycle and regenerative cycle configuration, however, the regenerative cycle



turbine using a conventional combustor will have much higher NO  emissions.
                                                               X


Examples of this are depicted in Figures 3-50  and 3-51, which show NO  emissions
                                                                     A


versus firing temperature for a General Electric Model 7001 B gas turbine



burning distillate oil and used in the simple  cycle and regenerative cycle



configurations, respectively.  The much greater emission of NO  from the
                                                              A
                                   3-104

-------
regenerative cycle gas turbine 1s due to the Increased combustor Inlet



temperature and the Increased residence time of the combustion gases at high



temperatures as described 1n section 3.2.3.4.1.



     Figures 3-50 and 3-51 also depict the asympotic relationship between NO
                                                                            A


emissions and the firing temperature (which is directly related to combustion



temperature) of the turbine.  Since firing temperature is directly related to



the power output of the gas turbine, Figures 3-50 and 3-51 show that NO
                                                                       A


emissions increase with increasing gas turbine load.



3.2.3.5  State and Local Regulations for Statjonary Gas Jurbjnes  -  Only a



few state and local jurisdictions have regulations specifically for stationary



gas turbines, although a considerable number have applied standards originally



written for other sources to gas turbines.   '      A summary of these



regulations is shown in Table 3-10.  Thus, the most stringent state regulation



on NO  from fuel combustion, which could be applicable to gas turbines of all
     /\


sizes, is 0.2 Ib NO /million Btu for gas fired burners and 0.3 Ib NO /million
                   X                                                X


Btu for oil fired burners.  Eight states have this standard.  San Diego County



appears to have the most stringent NO  regulations limiting emissions to 75 ppm
                                     X


(about 0.3 Ib NO /million Btu) and 42 ppm at 15 percent oxygen when burning
                /\

                                       178
liquid and gaseous fuels, respectively.



     State sulfur regulations vary from 0.3 to 2.6 weight percent with an


                           179
average of about 1 percent.     State visible emission regulations vary from



0 to 60 percent opacity with the typical standard, applicable to gas turbines


                                                             180
in 40 states,  restricting opacity to "less than 20 percent".     There are no



known state regulations for CO emissions from gas turbines.



     Table 3-11 is a tabulation of the typical and most stringent state or local



standards to which gas turbines are subject.  Table 3-12 shows the effect of such



standards on emissions of NOX and S02 from a typical large (33 megawatt) gas turbine.
                                    3-105

-------
D-
OL-
190

180


170


160


150


140


130.


120

110


100


90


80

70

60


 50


 40


 30

 20


 10
                 Figure 3-5(r   NO  Emissions vs. Firing Temperature for an
                               MSX7001-B Simple Cycle Turbine Burning Distillate
                               Fuel  (I.S.O. Conditions)
                                                             Peak
                                                             Load
                                                        Base
                                                        Load
                1000
                      1200
1400
1600
1800
2000
2200
                               Firing Temperature -  F

                                   3-106

-------
0-
O-
380


360 |	


340


320


300


280


260


240


220


200


180


160

140


120


100


 80


 60


 40


 20
                             174
                  Figure  3-51      NOX Emissions vs.  Firing  Temperature  for  an
                                  MS 7001-B Regenerative Cycle Turbine  Burning
                                  Distillate Fuel  (I.S.O.  Conditions)

                                                                   Base
                                                                   Load
                800
                     1000
1200
1400
1600
1800
2000
                             Firing Temperature -  F
                                   3-107

-------
Table 3-10.  Summary of State Regulations
                                         177

Alabama '
Alaska
Arizona
Arkansas
California
Colorada
Connecticut
Delawara
Disc, of Col.
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusatta
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
N. Hampshire
New Jersey
New Mexico
New York
N. Carolina
N. Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
S. Carolina
S. Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Samoa
Guam
Puerto Rico
Virgin Islands
UEL BURNING LBS NO /106 BTU
OF HEAT INPUT/HR
OA-S
0.2





0.2
0.2
0.2
0.2
0.2


0.3
0.2
0.2
0.3
0.2


0.2
0.3

0.2



0.2


0.213,14
0.2
0.6

0.2
0.2

0.3
0.27

0.2


« a
IB
0.2


0.2
0.2




LIQUID | SOLID
-0.3





0.3
0.3
0.3
0.3
0.3


0.3
0.3
0.3
0.3
0.3


0.3
0.3

0.3



0.3


0.313
0.3
0.6

0.3
0.3

0.3
0.3?

0.3



0.3 	
0.3


0.3
0.3




0.7





0.75
0.3
0.7
0.7
0.7


0.7
0.7


0.7


0.5
0.3








0.452'15
0.7
1.30

0.9
0.7









0.7


0.7





                  3-108

-------
oo
oo
                 C3
                     ra
                     E    3   E
                     Q.         Q.
                     ca         a:
                                          9
                                3-109

-------

         .    _.
       — 1   
^
                    —
                    co
                    co

                             °°
         S
                                    £
                                            ifi
                        •=r
                                         O  i — I
                               s
s
                                                CM
                            r- (=1
                            1— >—<


                            CO ^5
                            '1— •—«
                            o —I
      S
                                             3-110

-------
souvce

 6.

                           ,75.
                          *"

                           • -


                                                            l



-------

""
        World c   "  "*"1""le''». F
                             r

                               '
       „
     »  "•
            . Jo
               '
          * • and



    F
                    9-)0.

                              .   „„

                             T  „  Cent
                      ,,



                           .
.   ,.  ;;ce H-  > ..

         '   e"


                              "t

-------

-------
rA
      g
      *—t
      ee





      R
         CO
                 ca
                     C3
                ^
                     ca
                                ca
                      £    3
                      Q_


                     9"   2
                           3
ce

£
a.


K
                                      CC
           3
                                      oS
                                      ca:
ff

 3
CO
83

      i
      CO
                                                      1 1 [
                                                o    52
                                                a    >
                                3-109

-------
a
s   S
       I
I
oo
       &5   S

       S   S
       ii    n"l


       d   £
            £
            R      s
            ^
            
-------
                         REFERENCES FOR CHAPTER 3

1.  Report prepared by the Tennessee Valley Authority, "Analysis of the
    Cost and Benefits of Nitrogen Oxide Emission Control on Proposed
    Johnsonville and Gallatin Gas Turbines".  November 1974.
2.  Preliminary draft report prepared for EPA by Energy and Environmental
    Analysis, Inc.  Contract No. 68-02-2082.  "Economic Impact of New
    Source Performance Standards on Stationary Gas Turbines",  pp 1-3
    through 1-40.  December 2,,1975.
3.  Tennessee Valley Authority, Reference 1.
4.  De Biasi Victor.   "^ 50 Design Shortcut to 1980 Technology".  Gas
    Turbine World,  pp 9-17.  November 1975.
5.  Stambler, Irwin.  "EPRI Gas Turbine Outlook".  Gas Turbine World.
    pp 19-22.  November 1975.
6.  Report No.  SRL 1378-01-0374 prepared by Scott Research Laboratories,
    Inc., for General Applied Science Laboratories as an Appendix to
    GA SL-TR-787.  "Turbine Exhaust Emissions Measured at Facilities of the
    New York Power Pool",  pp 9-28.  March 1974.
7.  Johnson, R. H. and M. B. Hilt.  "Gas Turbine Environmental Factors - 1971".
    General Electric Publication GER-2486.  p 5.
8.  Lipfert, F. W.  "Correlation of Gas Turbine Emissions Data".  General
    Applied Science Laboratories.  ASME, Publication 72-GT-60.  p 3.  1972.
9.  Dibelius, N. R., and E. W. Zeltmann. "Gas Turbine Environmental Impact
    Using Natural Gas and Distillate Fuel".  General Electric publication
    73-CTD-6.  p 40.  February 1973.
                                    3-111

-------
10.   Johnson, R.  H., and Wilhelm,  F.  C.,  "Control  of Gas Turbine Emissions 1n
     the World Environment".   General Electric Company,   p 9.   1974.
11.   Ibid, Reference 10. pp 9-10.
12.   Ambrose, M.  J., and E. S. Obidinski.   "Recent Field Tests for Control of
     Exhaust Emissions from a 35 MW Gas Turbine".   Westinghouse.  ASME publication
     72-JPG-GT 2.  p 5.  September 1972.
13.   Letter and attachments from Eugene W. Zeltman, General Electric, to
     Kenneth R. Durkee, ESED, EPA.  pp 52-67.  October 31, 1975.
14.   Lipfert, F.  W., and E. A. Sanlorenzo, and H.  W. Blakeslee.  "The New York
     Power Pool Gas Turbine Emissions Test Program".  Paper presented at  the
     APCA Specialty Conference,  p 6.  October 13 - 15, 1975.
15.   Ibid, Reference 14.  p 6.
16.   Op. Cit., Reference 12, p. 10
17.   Johnson, R.  H., and C. Wilkes.  "Environmental Performance of  Industrial
     Gas Turbines".  General Electric.  ASME publication 74-GT-23.  p 6.
     April 4, 1974.
18.  Carl, D. E., and  E. J. Obidinski, and C. A.  Jersey.   "Exhaust  Emissions
     from a  25 MW Gas  Turbine  Firing, Heavy  and Light Distillate Fuel Oils
     and Natural Gas".  Westinghouse Electric Corporation.  ASME Publication
     75-GT-68.   pp  8-11.   December 2, 1974.
19.  Appendix  to GASL-TR-787  ("Gas Turbine Emission Measurement Program"
     for Empire  State  Electric Energy Research Corporation, New York, N.Y.).
     "Turbine  Exhaust  Emissions Measured  at  Facilities  of  the  New York  Power
     Pool'.   Prepared  by Scott Research Laboratories, Inc.  pp 9-26.  March  1974.
9.0.  Ibid, Reference  19.   pp  9 and 34.
21.  Op. Cit., Reference" f4,~"p". "4-~5~.
22.  Op. Cit., Reference 18,  p. 8-9.
                                   3-112

-------
23.  Or. Cit., Reference 10, p. 17.
?<.  Op. Cit., Reference 14, p. 5.
25.  Bahr,  Donald W.  "Technology for the Reduction of Aircraft Turbine
     Engine Exhaust Emissions".  General Electric.  Presented at the
     Industry-Military Jet Fuel Quality Symposium, San Antonio, Texas.
     p 9.  January 30 - February 1, 1973.
26.  Op. Cit., Reference 13, p. 62-64.
27.  General Electric Corporation, Brochure GEA-9679.  "General Electric
     Heavy-Duty Gas Turbines 5002".  p 5.
28.  Strong, R. E., and C.  E. Hussy.  "Combustion-System Design for
     Industrial Gas Turbines".   Westinghouse.  ASME Paper Number 64-GTP-2.
     p 2.  November 1964.
29.  Dilmore, J.  A., and W. Rohrer.  "Nitrogen Oxide Formation in the
     Combustion of Fuels Containing Nitrogen in a Gas Turbine Combustor".
     ASME 74-GT-37. p 5.  November 1973.
30.  "Advanced Combustion Systems for Stationary Gas Turbines".  Proposal
     prepared by General Electric in response to EPA, RFP Number Du-75-A182.
     p 8.  May 9, 1975.
31.  Op.  Cit., Reference 28, p. 2.
32.  Op.  Cit., Reference 30, p. 6-8.
33.  Saintsbury,  J. A., and P.  Sampath.  "Emissions Research on a Simple
     Variable Geometry Gas Turbine Combustor".  Pratt & Whitney.  ASME
     Publication 75-WA/GT-12.  p 3.
34.  Stambler, I.  "EPRI Gas Turbine Outlook".  Gas Turbine World,  p 21.
     November 1975.
35.  Op. Cit., Reference 28, p.  1.
                                   3-113

-------
36.   Op. Cit.. Reference 30, p. 8.
37.   Op. Cit., Reference 17, p. 4.
38.   Op. C1t., Reference 10, p. 14.
39.   Op. C1t., Reference 17, p. 4.
40.   Op. Cit., Reference 17, p. 4.
41.   Op. Cit., Reference 10, p. 14.
42.   Decorso, S.  M., et al.   "Smokeless Combustion in Oil-Burning Gas Turbines".
     ASME Publication 67-PWR-5.  pp 1-2.   July 1967.
43.   Ibid, Reference 17.  pp 3-4.
44.   Ibid, Reference 10.  pp 14-16.
45.   Carl, D. E., et al.  "Exhaust Emissions from a 25 MW Gas Turbine Firing
     Heavy and Light Distillate Fuel Oils and Natural Gas".   ASME Publication
     75-GT-68.   p 9.  March 1975.
46.   Op. Cit., Reference 42, p. 1-2.
47.   Op. Cit., Reference 10, p. 14.
48.   Op. Cit., Reference 17, p. 3-4.
49.   Op. Cit., Reference 10, p. 15.
50.   Op. Cit., Reference 10, p. 15.
51.   OD. Cit., Reference 10, p. 16.
52.   Op. Cit., Reference 10, p. 16.
53.   Vaught, 0. M.  "The Effect of Inlet Temperature and Pressure on an Industrial
     Turbine Engine Exhaust Emission".  ASME Publication 75-WA/GT-ll.  p 7.
     July 28, 1975.
54.   Op. Cit., Reference 42, p. 2.
55.   Op. Cit., Reference 25, p. 7.
56.   "General Motors Response to Preliminary Proposed Standards for Control of
     Air Pollution from Stationary Gas Turbines".  Submitted to EPA, OAQPS, ESED.
     p 34.   March 21, 1973.
                                      3-114

-------
57.   Op.  Cit., Reference 42, p.  2.
58.   Wassell, A.  B.   "Development of Pollution Controls for Rolls-Royce
     RB 211  and Olympus 593 Engines."  SAE Publication 740843.   pp.  1-6.
     April  1974.
59.   Op.  Cit., Reference 42, p.  2.
60.   Op.  Cit., Reference 58, pp.  1-6.
61.   Op.  Cit., Reference 17, p.  7.
62.   Op.  Cit., Reference 25, p.  7.
63.   Op.  Cit., Reference 56, p.  34.
64.   Niedzwiecki, R. VI., and E.  J.  Jones,  "The Experimental Clean Combustor
     Program - Description and Status."  Lewis Research Center.   NASA Technical
     Memorandum, NASA-TM-X-71547.  NTIS Number N74-21399.  p. 5.   April 1974.
65.   Op.  Cit., Reference 8, p. 3.
66.   Op.  Cit., Reference 25, pp.  13-21.
67.   Op.  Cit., Reference 25, p.  13.
68.   Op.  Cit., Reference 64, p.  5.
69.   Op.  Cit., Reference 10, p.  11.
70.   Op.  Cit., Reference 10, p.  10.
71.   Op.  Cit., Reference 64, p.  5.
72.   Op.  Cit., Reference 25, pp.  13-15.
73.   Op.  Cit., Reference 17, pp.  4-5.
74.   Op.  Cit., Reference 8, p. 3.
75.   Op.  Cit., Reference 25, p.  17.
76.   Op.  Cit., Reference 25, p.  115.
77.   Op.  Cit., Reference 10, p.  10.
78.   Op.  Cit., Reference 25, p.  18.
79.   Op.  Cit., Reference 53, p.  6.
                                      3-115

-------
80.   Op. CH. , Reference 53, p. 7.
81.   Op. C1t., Reference 18, p. 8.
82.   OD. Cit., Reference 18, p. 8.
83.   Op. Cit., Reference 53, p. 8.
84.   Klapatch, R. D.   "Gas Turbine Emissions and Performance on Methanol Fuel".
     T'-rbo Power and Marine.  ASME Publication 75-PWR-22.  pp 1-4.  June 30,
     1974.
85.   Pillsbury, P. W., and S. S. Lin.  "Parametric Tests on Dual-Fuel Coal
     Gas/Distillate Oil Gas Turbine Combustors".  Westinghouse Electric
     Corporation.  ASME Publication 75-PWR-13.  pp 1-11.  June 23, 1975.
86.   Ibid, Reference 85.  p 6.
87.   Op. Cit., Reference 85, p. 6.
88.   Hung, W. S. Y.  "An Experimentally Verified NO  Emission Model for Gas
                                                   J\
     Turbine Combustors".  ASME Publication 75-GT-71.  pp 1-11.  December 1974.
89.   Dilmore, J. A., and W. Rohrer.  "Nitric Oxide Formation in the Combustion
     of Fuels Containing Nitrogen in a Gas Turbine Combustor".  ASME 74-GT-37.
     pp 1-11.  April 1974.
90.   Sarli, V. J., et al.   "Effects of Operating Variables on Gaseous Emissions"
     Presented at the APCA  Specialty Conference on air pollution measurement
     accuracy as it relates to  regulation compliance,  pp 14-17.  October 26-28,
     1975.
91.   Letter and attachments from H. Gaylord, Turbodyne Corporation, to Don  R.
     Goodwin, EPA, ESED.  Section G.3.4.  pp 1-6.  December 19, 1975.
92.   Op.  Cit., Reference 89, pp. 1-11.
°'3.   OP.  Cit., Reference 91. pp. 1-6.
94.   Op.  Cit.,  Reference 91, "pp~.  1-6.
                                    3-116

-------
 95.   Op.  Cit., Reference 89, p. 3.



 96.   Op.  Cit., Reference 91, pp.  1-6.



 97.   Op.  Cit., Reference 90, pp. .14-17.



 98.   Op.  Cit., Reference 89, p. 3.



 99.   Op.  Cit., Reference 25, pp.  25-26.



100.   Op.  Cit., Reference 25, pp.  25-26.



101.   Bahr, D. W., and C. C. Gleason.  "Technology for the Reduction of Aircraft



      Turbine Engine Pollutant Emissions".  ICAS Paper Number 74-31.  pp 8-9.



      August 25-30, 1974.



102.   Op,  Cit., Reference 91, Section G.3.3., p. 2-4.



103.   Op.  Cit., Reference 30, p. 15.



104.   Blazowski,  W. S., et a!.  "Prediction of Aircraft Gas Turbine NO  Emission
                                                                      /\


      Dependence on Engine Operating Parameters and Ambient Conditions".  AIAA



      Paper Number 73-1275.   p 1.  November 1973.



105.   Ibid, Reference 104.  p 6.



106.   Vermes, G.   "A NO  Correlation Method for Gas Turbine Combustors Based on
                       /\


      NO  Formation Assumptions".  ASME Publication 74-WA/GT-10.  p 5.  August
        A


      1974.



107.   Sullivan, D. A., and P. A. Mas.  "A Critical Review of NO  Correlations
                                                               ^


      for Gas Turbine Combustors".   ASME Publication 75-WA/GT-7.  pp 1-12.



      July 28, 1975.



108.   Letter and attachments from Dibelius, N. R., Chairman, ASME Combustion and



      Fuels Committee to Don R.  Goodwin, ESED, EPA.  May 12, 1975.



109.   Letter and attachments from McGaw, W. L., Chairman, American  National



      Standards Institute B133.9 Task Force, to Don R. Goodwin, EPA, ESED.



      January 9,  1975.



110.   Letter and Enclosures  from Zeltman, E., General Electric Corporation, to



      Kenneth R.  Durkee, EPA, ESED.  September 4, 1975.




                                     3-117

-------
111.   Appendix 21  of letter from Assard,  D.  G.,  Turbo  Power  and Marine,  to
      Don R.  Goodwin, ^PA,  ESED.   November 26,  1975.
112.   Letter and attachments from Decorso, S. M.,  Westlnghouse, to  Don  R.
      Goodwin, EPA, ESED.   pp 30-36.   January 8,  1976.
113.   Vaught, J. M.  "The  Effect of Inlet Temperature  and Pressure  on an
      Industrial Turbine Engine Exhaust Emissions".  ASME Publication
      7B-WA/GT-11.  p 3.  July 28, 1975.
114.   Letter and attachments from Sievert, M. 0.,  President, Solar, to  Don  R.
      Goodwin, EPA, ESED.   November 10, 1975.
115.   Letter and attachments from Medigovich,  D.  G., Garrett Research,  to
      Don R.  Goodwin, EPA,  ESED.   pp 2-3, 5-9.   February 13, 1976.
116.   Letter and attachments from Gaylord, R.  H.,  Turbodyne  Corporation, to
      Don R.  Goodwin, EPA,  ESED.   Section G-l.   December 19, 1975.
117.   Op. Cit., Reference  107, p. 8.
118.   Touchton, G. L., and N. R.  Dibelius.  "A Correlation of Nitrogen  Oxides
      Emissions with Gas Turbine Operating Parameters".   ASME Publication
      76-GT-14.  pp 2-3.  March 21, 1976.
119.   Ibid, Reference 118.   pp 2-3.
120.   Op. Cit., Reference  118, p. 4.
121.   Bahr, D. W., and C.  C. Gleason.  "Experimental  Clean Combustor Program",
      Phase I Final Report Number NASACR 134737 prepared for the National
      Aeronautics and Space Administration,  p 55.  June 1975.
122.   Op. Cit., Reference  104, pp. 5-6.
123.   OD. Cit., Reference  107, pp. 6-7.
12*.   op. Cit., Reference  107, pp. 4-5.
125.   Op. Cit., Reference  113, p. 7.
126.   Op. Cit., Reference  113, p. 10.
                                     3-118

-------
127.  Op.  Cit., Reference 114.
128.  Op.  Cit., Reference 112.
129.  Op.  Cit., Reference 91, Section 6-1.
130.  Op.  Cit., Reference 121,  p.  55.
131.  Op.  Cit., Reference 116,  Section G.3.3.,  pp.  1-10.
132.  Op.  Cit., Reference 121,  pp.  55.
133.  Op.  Cit., Reference 121,  pp.  55.
134.  Op.  Cit., Reference 10, p.  11.
135.  Op.  Cit., Reference 30, pp.  18-24.
136.  Op.  Cit., Reference 116,  p.  3-9.
137.  Op.  Cit., Reference 29, p.  1-10.
138.  Op.  Cit., Reference 29, p.  2.
139.  Op.  Cit., Reference 10, p.  11.
140.  Op.  Cit., Reference 30, pp.  18-19.
141.  Op.  Cit., Reference 10, pp.  11.
142.  Op.  Cit., Reference 30, p.  19.
143.  Op.  Cit., Reference 10, p.  11.
144.  Op.  Cit., Reference 30, p.  19.
145.  General  Electric submittal  to EPA based on  the work of R. Johnson and
      C. Wilkes, and edited by  H.  Hamilton  and  E. Zeltman.  "Comments of the
      General  Electric Gas Turbine Products Division on  the Impact of Fuel-
      Bound Nitrogen on the Formation of Oxides of  Nitrogen from Gas Turbines",
      pp.  1-25.  January 28, 1974.
146.  Ibid, Reference 145, p. 5.
147.  Op.  Cit., Reference 89, p.  9.
148.  Op.  Cit., Reference 30, pp.  19-20.
149.  Op.  Cit., Reference 30, p.  19.
                                     3-119

-------
150.   Op. Cit., Reference 91, Section 6.3.3., pp.  1-10.
151.   Decorso, S.  M.  "Recent Developments in the  Use of Heavy Fuels in
      Industrial  Gas Turbines".   Westinghouse.   Presented at the 9th
      International  Congress on  Combustion Engines,  Stockholm, Sweden.
      1971.   p 9.
152.   0.. Cit., Reference 30, p.  23.
153.   Hung,  W. S.  Y.  "The Reduction  of NO  Emissions from Industrial  Gas
                                          J\
      Turbines",   llth International  Congress on Combustion Engines,
      Barcelona.   1975.   p 6.
154.   Op. Cit., Reference 18, p.  7.
155.   Supplied to EPA by Eugene  W. Zeltmann.  General Electric Corporation.
156.   Ibid,  Reference 84.  p 3.
157.   Letter and attachments from Zeltmann, E.  W., General Electric Corporation,
      to Kenneth R.  Durkee, EPA, ESED.  August 27, 1975.
158.   Op. Cit., Reference 85, p.  6.
159.   Op. cit., Reference 85, p.  6.
160.   Op. Cit., Reference 85, p.  6.
161.   Op. Cit., Reference 85, p.  6.
162.   Schieffer, R.  B., and D. A. Sullivan.  "Low Btu Fuels for Gas Turbines".
      ASME Publication 74-GT-21.  pp 1-8.  November 13, 1973.
163.   Crouch, Schlinger, Klapatch, and Vitti.  "Recent Experimental Results on
      Gasification and Combustion of Low Btu Gas for Gas Turbines".  ASME
      Publication 74-GT-ll.  p 10.  November 3, 1973.
164.   Pillsbury, et al.  "Emission Results from Coal Gas Burning in Gas Turbine
      Combustors".  ASME Publication 75-GT-44.   p 7.  March 6, 1975.
165.
                                     3-120

-------
166.   DeDiasi, V.   "Gas Turbine Utility Business Will  Never Be The Same Again".
      GfL§ Turbine  World,   p 22.  May 1975.
167.   Sternlicht,  B.   "The Equipment Side of Low Level Heat Recovery".  Power.
      Volume 119.   Number 6.  p 72.  June 1975.
168.   Armstrong, C.  H.  "Technology Advances Raising Combined Cycle Efficiency".
      Gas Turbine  International,  pp 38-41.  September - October 1974.
169.   Ibid, Reference 168.
170.   Op. Cit., Reference 168.
171.   Op. Cit., Reference 30,  p. 59.
172.   OD. Cit., Reference 30,  p. 61.
173.   Supplied by  E.  Zerltmann, General Electric Corporation.
174.   Op. Cit., Reference 171.
175.   Dibelius, N. R., and R.  J. Ketterer.  "Status of State Air Emission
      Regulations  Affecting Gas Turbines".  General Electric.  ASME Publication
      73-WA/GT-8.   1973.
176.   Duncan, L. J.   "Analysis of Final State Implementation Plans - Rules and
      Regulations".   APTD-1334.   pp 8, 64-65T
177.   Ibid, Reference 176.  pp 64-65.
178.   "Rules and Regulations".  Rule 68 (p 33) and Rules 50, 52, and 62 (pp 19,
      20, and 26,  respectively).  San Diego County Air Pollution Control
      District.  February 1972.
179.   OD.  Cit., Reference 176, pp.  58-63.
180.   Op.  Cit., Reference 176, pp.  51-52.
                                      3-121

-------
                      4. EMISSION CONTROL TECHNIQUES
4.1  PARTICULATE EMISSIONS
4.1.1  General
       As discussed in sections 3.2.2.1 and 3.2.2.2, participate emissions
from gas turbines consist of ash from the fuel and particulates of carbon and
hydrocarbons resulting from incomplete combustion.  Fuels containing high
ash and vanadium contents will result in higher parti oil ate emission rates
than light distillate fuels or natural gas.  As discussed in section 3.2.2.1,
particulate emissions from a turbine operating at 20 MW and burning natural
gas were about 0.2 Ib/MWH compared to 0.8 Ib/MWH for, a.turbine operating at
52 MW and burning #2 oil.  Because of the high gas velocities, extreme turbulence
caused by the ducting configuration and sound baffles, and the difficulty
of sampling with very long probes in the environment of the exhaust gases
(as discussed in section 3.2.2.1), measurement of particulate emissions from
gas turbines are of questionable accuracy and high variability.  Magnesium,
manganese and barium inhibitors added to liquid fuels to retard  vanadium
                                                                             123
corrosion of the turbine components will also increase particulate emissions.
4.1.2  fipntrol Techniques for Particulatp Fmissions from Gas Turbines
       Particulates are emitted from gas turbines at low levels;  varying
from 0.002 gr/scf to 0.10 gr/scf and exhaust gas volumes hanging from 166,206 scfm
to 460.167 scfm for turbines operating at base loads of 12 and 44
megawatts, respectively. ' ' '
       As discussed in sections 3.2.2.1, 3.2.2.2, and 3.2.2.3, particulate
emissions may be decreased by burning natural gas or low ash fuels and by
combustor modifications which provide more complete combustion of hydrocarbons
and carbonaceous particles.
       The only known application of a particulate control  device for gas
turbines has been the installation of a packed bed scrubber on several jet-
                                     4-1

-------
engine test cells at a naval  air station.   This scrubber reportedly reduced
particulate emissions from uncontrolled levels ranging  from about  0.008 gr/scf
to 0.015 gr/scf, dep',.ding on engine operating mode, to controlled levels
                                          8 9
ranging from 0.001 gr/scf to 0.005 gr/scf.  '   Opacity  after the scrubber
was generally less than 10%.
       A jet engine test cell is not typical  of a stationary gas turbine
installation.  Because a jet engine operates in different modes, including
an afterburner mode, opacities vary from 0 to 80 percent whereas a typical
stationary gas turbine will operate with opacities generally less  than 20
percent.  The scrubber discussed above was installed to reduce the opacity,
noise and particulate emissions to meet requirements for Federal facilities.
A brief discussion of the costs which would be incurred to install such a
scrubber on stationary gas turbines is presented in Chapter 7.
4.2  Visible Emissions
       The factors which affect visible emissions have  been discussed in section
3.2.2.2.1.  Visible emissions are related to particulate emissions so the
factors discussed in section 4.1 to reduce particulate  emissions from stationary
gas turbines may also reduce visible emissions.
       One way to reduce visible emissions is to burn fuels with high
hydrogen contents.  This is shown by the data in figures 3.13 and 3.14.
Another method for reducing visible emissions is to use fuel additives such
as soluble compounds of manganese, barium, lead, iron and others.   The
actual mechanism by which these additives reduce visible emissions is unknown
and somewhat controversial.  For example, a paper by personnel of Ethyl Corporation
states that the mass of particulate emissions is reduced by the use of manganese
additives while a paper by personnel of General Electric does not support such
a conclusion.  '
                                       4-2

-------
       Major  reductions  1n visible emissions have been achieved through
combustor  redesign.  The design of low visible emission combustors involves
providing  leaner fuel-to-air mixtures and more affective fuel and air mixing
in the primary zone.   '  '    Figure 4.1 shows visible emission data for
General Electric aircraft turbines developed both prior to 1966 and after 1966.
The figure shows that  visible emissions from engines developed after 1966,
which incorporate combustor design modifications to reduce visible emissions, are
below the threshold of visibility.  The newer models of stationary gas
turbines have also incorporated combustor modifications such that visible
emissions are below the  visible threshold and/or are below 10 percent opacity
for turbines burning fuels ranging from distillates to residuals.     '  '
Figure 4.2 shows the relative effect of using a fuel additive versus combustor
modifications to reduce  visible emissions.  Visible emissions increase
exponentially with Bacharach number (e.g. to reduce visible emissions from
Bacharach 9 to 4 requires eight times the smoke reduction as that to reduce
                                         20
visible emissions from Bacharach 5 to 4.)    Figure 4.2 shows that combustor
modifications reduced  emissions substantially while the use of the manganese
additive had relatively  little effect.
4.3  S02 Emissions
       As stated in section 3.2.2.1, S0£ emissions from gas turbines are strictly
a function of the fuel sulfur content and virtually all fuel sulfur is converted
to SOp.   The only technique now used to control SOp emissions from gas turbines
is to burn low sulfur  fuels.   Stack gas scrubbing for SOp removal has not been
applied to gas turbines  primarily because it is less costly to desulfurize
the fuel.   Relative costs are briefly presented in Chapter 7.  Figure 4.3 shows
emission rates from gas  turbines for various fuel sulfur contents and for
varying  fuel  flow rates.
       Distillate fuels  range in sulfur content from 0.01  to 0.48 percent by

                                      4-3

-------
                Figure  4.1
                            15
        Comparison Of Peak Engine  Smoke Emission Characteristics Of Various
        General  Electric Aircraft  Turbine Engines, At Standard Day-Sea Level
        Static Operating Conditions
   100
    80
    60
     40
     20
                                                      Engines Developed
                                                      Prior  to 1966
                                       Small  Engines
                      Approximate Visibility Threshold
              Large Engines
                                Engines Developed Since 1966
                                	1	I
                           10        15        20        25

                          Engine Cycle Pressure Ratio Rating
Note:  Low smoke numbers  indicate  low visible  emission  levels.
                                       4-4

-------
Figure 4.221

The Effect  of  Fuel  Additives and  Combustor Redesign on
Visible Emissions
  CD
  X
  o

  OB
           FULL LOAD OPERATION
           25   50    75   100   125    ISO


           PP M MANGANESE IN *2 DISTILLATE FUEL
 Note:  Bacharach  smoke number decreases with decreasing
        visible emissions.
                          4-5

-------
              21
  Figure 4,3


  SCL  Emission  Rate as  a Function  of  Fuel  Sulfur

  Content and Gas  Turbine Fuel Flowrate
              SO, EMISSION RATE - KGWR
            0.038 0.02
   80   60

SO, EMISSION

RATE - KG/HR
  5


-20
                    -to
10 15 20  25

TOTAL FUEL FLOW

RATE - KG/SEC
               '        H2SO« 2H2O EMISSION


                  "*"*   RATE-"KG/HR
                     4-6

-------
weight; crudes contain from 0.06 to 3.0 percent by weight; and residual oils
                                       22 23
vary from 0.5 to 3.2 percent by weight.  *    Gas turbines have historically
burned the low sulfur distillate fuels or natural gas.  During the early 1970's
there was a trend towards the installation of fuel treatment facilities to treat
residual and crude oils so they could be fired in gas turbines.  These treatment
facilities removed the salts from the fuels and added inhibitors to prevent
the vanadium in the fuel from corroding the gas turbine components but did not
reduce the sulfur content.  This trend toward the use of residual and crude
oils is not evident today and the use of treatment facilities already installed
                 24 25 26 27
has been minimal.  '*       In a September 1975 survey of electric utilities,
only four of 48 respondees indicated plans to burn residual oil in their gas
         28
turbines.    In response to a questionnaire distributed to electric utilities
by Edison Electric Institute in October 1975, 17 of 17 respondees indicated
                                                                            29
that no trend existed to use crude or residual oils for firing gas turbines.
Therefore, it appears that the low sulfur distillate oils will continue to be
burned by the owners and operators of gas turbines as a means of controlling S02
emissions and that high sulfur crude oils and residual oils will not be used.
The primary reason for the shift away from crude and residual fuels is one
of economics.  In todays market, it simply costs more to buy and treat the crude
                                                                30
and residual oils than to purchase and burn the distillate oils.
       In the event that the trend to crude and residual oils again emerges,
the control technique for reducing S02 emissions is to remove the sulfur from
the oil prior to its combustion in the turbine.
4.4  HYDROCARBON AND CARBON MONOXIDE EMISSIONS
4.4.1   General
       As shown in figures 3.17 and 3.18, hydrocarbon and carbon monoxide emissions
from conventional gas turbines are low at full load operation and increase
asymptotically at low turbine loads.  In section 3.2.2.3.1, it has been shown
that incomplete combustion is the principal cause of hydrocarbon and carbon
                                      4-7

-------
monoxide emissions from gas turbines.   Most existing turbines  are  designed
for maximum efficiency at full  load operation with efficiency  dropping  to
90-95 percent at low Irid operation as discussed in 3.2.2.3.1.
       Gas turbines with higher combustion efficiencies  tend to have lower
CO emissions as can be observed by comparing the heat rates (inverse of
efficiency) and CO emissions for the turbines shown in Table 3.5.   The  CO
emission1, versus turbine size for small and large gas turbines without  NO
                                                                         A
controls are shown in Figures 4.4 and 4.5, respectively.  The  data show
that no generalization can be made regarding CO emissions versus turbine size,
It is generally accepted as a "rule of thumb" that, for large  gas  turbines,
a 1 percent decrease in turbine efficiency is incurred for each 300 ppm of
             31
CO emissions.    HC emissions also increase with decreasing combustion
efficiency.  Figures 4.6 and 4.7 show HC emissions versus turbine  size
for small and large turbines, respectively.  These figures show that most
gas turbines, regardless of size, have HC emissions generally  less than 20 ppm
when operated at or near full load.  As explained in section 3.2.2.3.1,
combustion chemical kinetics data show that vaporized hydrocarbons and  any
partially oxidized hydrocarbons are consumed during combustion much more
rapidly than CO and thus the major product of inefficiency is expected to be
                                 ita
                                 33
   op
CO.    This is demonstrated by data from an Airesearch GTC 85-90 gas turbine
used in ground cart applications.'
        Turbine load   horsepower        0        121       168       192
        Combustion efficiency   %      97.2        98.6      99.3      99.6
        Hydrocarbons as CH4  Ibs/hr     1.576       0.627     0.141     0.061
        HC   ppm  @  15% 02            495        134        26        10
        Carbon Monoxide  Ibs/hr         7.37        7.21      5.38      3.34
        CO   ppm  @  15% 02           1324        880        566      315
                                      4-8

-------
   700
  600
  500
  400
   300
  200
C/9

o
   100
   50
O


O
                                                Q NATURAL GAS

                                                O LIQUID FUEL
                         0.5
                            1                     2

                          RATED POWER OUTPUT, Mw
   Figure 4.4.  CO emissions vs turbine size for small gas turbines without NOx controls when operated at
   or near full  load.
                                                4-9

-------
   200
CM


O
o.

in
   100
1   •
o
o
              D



               O
O


O
a
o


8
                                                               Q NATURAL GAS



                                                               O LIQUID FUEL
                                                                                  O



                                                                            0  lino
             10      15      20       25      30       35


                                          RATED OUTPUT,Mw
                     40
                    45
SO
55
60
   Figure 4.5. CO emissions vs turbine size for large gas turbines without NOX controls when operated at

   or near full load.
                                               4-10

-------
   18



   16



   14




   12




   10
CM  8
O
V)
z
O
SJ
X
                        T
                         O

                         O

                         O
             o
             O
                                                           Q  NATURAL GAS

                                                           Q  LIQUID FUEL

                                                               FUEL UNKNOWN
                                                                    a
                                                                    o

                                                                    o
                                                                    o

                                                                    o
                         0.5
                                        1                    2

                                         RATED POWER OUTPUT, Mw
   Figure 4.6.  HC emissions vs turbine size for small gas turbines without NOX controls when operated at
   or near full load.
                                               4-11

-------
   80
   70
   60
   50
   40
CM 30
o
in
»••
i-
   20
   10
                 a NATURAL GAS

                 O LIQUID FUEL
                                                              a


                                                              o
                                                              a

                                                              o
                     10
  20               30

RATED POWER OUTPUT, Mw
40
60
SO
    Figure 4.7.  HC emissions vs turbine size for large gas turbines without NOx controls when operated at
    or near full load.
                                                 4-12

-------
These data again Illustrate the trend for CO and HC emissions to decrease
with increasing load and show that HC emissions are generally much lower
than CO emissions.
4.4.2  Control of HC and CO Emissions
       Since combustion efficiencies are high for turbines operating at base
load conditions, HC and CO emission levels from stationary gas turbines are
usually very low as noted in Chapter 3.  As explained in section 3.2.2.3.1,
the control of HC and CO emissions from stationary gas turbines, therefore,
normally becomes important only at low loads where combustion efficiency decreases
for gas turbines using conventional combustors.  At low loads; low combustion
inlet air temperatures cause quenching, low fuel to air ratios result in low
equivalence ratios and reduced burning intensity in the primary zone, and
the low fuel and air flows result in poor fuel atomization and distribution.
       The control of HC and CO emissions is primarily a function of fuel
injection and atomization and fuel-air mixing.  Decreased HC and CO
emissions are therefore accomplished by combustor and fuel injection modifications
which promote better fuel atomization, better fuel and air mixing, and by controlling
the fuel to air ratios and residence time at temperature, as necessary to
provide combustion of the HC's in the primary zone of the combustor and
combustion of the CO in the primary and intermediate zones of the combustor.
       Unlike their stationary counterparts, gas turbines used in aircraft
are commonly operated at low power levels while the aircraft is on the ground.
Therefore, high levels of CO and HC are emitted at airports during these idle
conditions and the more recent efforts to control HC and CO have been targeted
at idle conditions.  These data are discussed to illustrate the reductions which
are achievable through application of combustor modifications.
                                      4-13

-------
       Figure 4.7 presents CO and HC emission data reported by General
Electric from tests of combustors for aircraft turbines.   The figure
demonstrates about a ?0 percent reduction in HC emissions and a 40 percent
reduction in CO emissions at idle power, when improved fuel atomization
techniques are applied.  Figure 4.8 shows that the CO and HC emissions  are
reduced by 75 and 90 percent, respectively, when fuel staging (sector burning)
techniques are applied at idle power to an annular combustor used in aircraft
turbines.  The figures shown here and in the remainder of this section  on HC and
                                                                I/ 21
CO emissions present these emissions as grams/kilograms of fuel.—  -
       Another method of reducing CO and HC emissions at low turbine loads without
adversely affecting turbine performance at high power conditions is to bleed off
a portion of the compressor discharge air before the combustor.  This controls
the fuel-to-air ratio in the combustor.  Figure 4.9 demonstrates that CO and HC
emissions from an annular combustor operating at idle conditions were reduced
by about 25 and 55 percent, respectively, when 20 percent of the compressor
discharge was bled overboard.
       An "Experimental Clean Combustor Program" funded by the National
Aeronautics Space Administration and performed separately by General Electric
and Pratt & Whitney has a primary objective to identify, define and develop
low emissions combustors for advanced aircraft engines.  A more complete
description of the combustor concepts tested is contained in section 4.5 - NO
                                                                             J\
Emissions.  Phase I of the effort by General Electric identified several
R & D combustor configurations which resulted in reduced HC and CO emissions.
FL.I staging at idle was a significant factor in all the design approaches.
Data for HC and CO emissions at idle conditions are presented in Figure 4.10 for
the major design approaches used in the investigation.  In Phase I, GE identified
-  ppm of CO at 15% 02 * 20 (Emission"index)
-/ ppm of HC as CH& at 15% 0^ « 32 (Emission Index)
                                      4-14

-------
       Figure 4.734


       CO  and HC  Emission Reductions with Improved Fuel Atomization  in

       an  Aircraft Gas Turbine at Ground Idle Power Operating Conditions
   80
«  60
h
bft
O
£
d
O
   40
6  20
                •  Fuel:  Kerosene
                Current Design
                     Modified Fuel  Nozzles

                     (All Primary Fuel Flow

                      At Idle)
                           ,-,  Airblast Fuel

                              Injection
                  CxHy

              (As Kerosene)
                                                     CO
                                 4-15

-------
       Figure 4.835
       CO and HC  Emissions Level  Reductions  1n an Aircraft  Gas Turbine
       at Ground  Idle  Power Conditions Using Various  Concepts of  Staged
       Fuel Injection
    ALL NOZZLES FUELED
                                              ALTERNATE NOZZLES FUELED
 OPPOSING SECTORS FUELED
                                                SINGLE SECTOR FUELED
    80
,2
S
§
    60
£   40
m
C
O
•H
a
ID
*•<
&
    20
O
o>
•
(H
O
                   •  Fuel:  Kerosene
               •  Current Design
                       Only Alternate Nozzles
                       Fueled At  Idle
                           • Only Nozzles  In
                             Opposing Sectors
                              Fueled At Idle
                                • Only Nozzles
                                  In A Single
                                  Sector
                                  Fueled At
                                      Idle
                   CxHy
               (As Kerosene)
                                                     CO
                                 4-16

-------
           36
Figure 4.9
CO and HC Emissions Level  Reductions in an Aircraft Gas Turbine,
at Ground Idle Power Operating Conditions, Standard Day-Sea Level
Static vs.  Compressor Bleed  Air Extraction
 024     6    8    10   12     14    16    18     20   22

 Compressor Bleed Air Extraction,  Percent Of Compressor Discharge Airflow
                          4-17

-------
       Figure 4.TO37

       CO and HC  Emissions  Reduction  Using Various Combustor Design
       Modifications  (combustor  rig tests)
   200
   100  -
    50
 01
 3
 bfi
 0)
•3
C
O
•H
m
in
•.-t
S
W

O
O
          Production Combustor (11-1)
 3
U,
 be
 X
 X
 bD
13
 O
 •f-4
 in
 tfl
 •H
 E
 W
                                1-4 Swirl-Can
                                Sector Burning
                       II-l Single Annular
                         Sector Burning  —
                         Rich Dome
                                II-9 Double Annular
                                  Outer Annulus
                                               II-7 Radial/Axial
                                                  Pilot Only
      0.005
0.010        0.015      0.020

         Metered Fuel-Air Ratio
0.025
0.030
                                4-18

-------
the double annular combustor and the radial/axial staged combustor as the two
most promising configurations for reduction of NO , CO and HC.  For the combustor
                                                 ^\
rig tests (data shown in figure 4.10), the radial/axial staged combustor
exhibited CO and HC reductions of about 58 and 93 percent, respectively, over
the current production combustor.  The double annular combustor exhibited
CO and HC reductions of about 49 and 78 percent, respectively, over the current
production combustor.
      Data for CO and HC emissions at idle conditions from the best
configurations of three design approaches investigated by Pratt & Whitney are
shown in Table 4.1.  This table shows that the swirl can configuration did not
result in reduced emissions as compared to the JT9D-7 production combustor.
The staged premix configuration P-3 reduced CO and HC idle emissions by 88 and
97 percent, respectively, as compared to the JT9D-7 production combustor.  The
swirl Vorbix configuration S3 did not significantly reduce CO or HC emissions,
but configuration S8 reduced CO and HC emissions by about 61 and 85 percent,
respectively, as compared to the JT9D-7 combustor.
      It must be stressed again that these combustors for which HC and CO
emission reductions have been discussed are designs to reduce the HC and CO
emissions at low loads for turbines which already have extremely low HC and CO
emissions at full load as demonstrated in Figures 4.11 and 4.12.  The data
illustrate, however, the viability of the control techniques.
4.4.3  Summary of Control Techniques for HC and  CO Emissions
      The data presented in this section show that combustion modifications
have been demonstrated which reduce CO emissions at idle power by 25 to 75
percent and HC emissions at idle power by 55 to  93 percent.  Turbines operating
at load have been shown to have minimal CO and HC emissions.
                                     4-19

-------
               Table 4.1
                         38
               SUMMARY OF CO  & HC EMISSIONS FOR  BEST CONFIGURATIONS OF
               EACH  COMBUSTOR CONCEPT  AT IDLE  CONDITIONS
 Combubtor Concept
 and Co; figuration

 Goals
         . reduction
 JT9D-7 Combustor
 Operating
 Condition

 With  and with-
 out bleed

 Without  bleed
 Swirl Can  Con figuration   With bleed
 N3 (Inner  row of
 carburetor  modules
 only)

 Swirl Can Configuration   Without bleed
 Nl 1 (Outer row  of
 carburetor  modules
 only)
Staged Premix
Configuration  P3

Staged Premix
Configurations P7.P8

Swirl  Vorbix
Configuration S3
With  bleed
Without bleed
With bleed
  Carbon
Monoxide

    20
Emission  Index (g/kg  fuel)
                     "Combustion
                       Efficiency
                       (Percent)
                       78.5
    Total
  Unburned
Hydrocarbons
                      58.5
     9                 1


       Unstable at design point


    68                29
                           99.1
77
117
29.8
61
94.7
90.1
                           91.3
                           99.8
                            95.0
Swirl  Vorbix
Configuration S8
Without bleed
    29
            4.5
                      98.8
Notes:    Combustor rig conditions with bleed were inlet pressure  of 2.93 atm,  inlet  temperature of 428 }
         and fuel-air  ratio  of 0.0126.

         Combustor rig conditions without bleed were inlet pressure of 3.74  atm,  inlet  temperature of
         456 K,  and fuel-air ratio of 0.0105.
                                            4-20

-------
   100
0)

fn
be   60
X
0)
•o
C
c
o
•H

-------
W)
X
0)
•a
0
•H
Ul
I/I
H
e
w

u
00
80
60
40
20
0





Q
A
•









\
\

GID
O







\
\
I








-N
\









> 	 r>









(
(





D No Bleed
) 3% CDP Bleed
JP-5 Fuel




CF6-50A
Standard Day
Power
30*
V
>
/•




Rated
•
859
^ r&




-
6 100%
1 O
400 500 600 700 800
                      Combustor Inlet Temperature,  c  K

                    40
         Figure 4.12H

         HC Emission Characteristics for the General  Electric CF6-50

         Production Engine/Combustor
                                     4-22

-------
4.5  NO  EMISSIONS
       /\


      As discussed In section 3.2.2.4, nitrogen oxides produced by combustion



of fuels 1n stationary gas turbines are formed by the combination of nitrogen



and oxygen 1n the combustion air ("thermal NO ") and from the combination of
                                             A


nitrogen in the fuel with oxygen from the combustion air ("organic NO ").  These
                                                                     /\


two forms of NO  contribute to the total NO  emissions from a given turbine.
               J\                           /\


The variables which affect NO  emissions from gas turbines have also been
                             X.


discussed in section 3.2.2.4.



4.5.1  Control of Thermal NO
         • i- I,  r    _- — L- - — •  _ .,-_  _j^


      It is generally recognized that there are four basic techniques to


                                                41  4?  4?  44  45  47  4ft
reduce the formation of thermal NO .  These are:*1' ^' w» HH' ™* *'• ™
                                  A


      a.  Reduce the combustion pressure



      b.  Decrease the peak flame temperatures in the combustor reaction



zone



      c.  Reduce the effective residence time during which the combustion gases



remain at elevated temperatures



      d.  Control the amounts of nitrogen and oxygen available for the



production of NO
                A


Most control techniques,  therefore, accomplish NO  reductions by designs which
                                                 rt


incorporate one or more of these techniques.



4.5.1.1  Control by Water or SteamInjection (Wet Techniques) - In section



3.2.2.4.1, it was shown that the formation of thermal NO  is extremely sensitive
                                                        /\


to flame temperature  . .  . increasing exponentially with increases in flame



temperatures.  Wet control techniques involve the injection of water or steam



into the combustion process.  The injected fluid provides a heat sink which



absorbs some of the heat  of reaction thereby reducing peak combustion temperatures



and the rate of NO  formation.
                  A
                                   4-23

-------
      The degree of NO  reduction achieved for a given turbine depends on
                      /\
the rate and method of introducing the water.   In experiments to maximize
effectiveness, manufacturers have found that direct injection of atomized
                                                     49  50  51  52
water into the primary zone of the combustor is best.  '              Figure 4.13
presents a summary of NO  emission data from gas turbines using wet control
                        A
techniques.  The effectiveness of wet control  techniques  in reducing NO
                                                                       /\
emissions is illustrated in Figure 4.14 where reductions  of NO  emissions
                                                              A
in excess of 80 percent are shown.  The band on Figure 4.14 is a compilation
of information provided by three major manufacturers of gas turbines, with the
spread being due primarily to differences in combustor design, type of fuel
used and method of water or steam injection.  Data points superimposed on the
band represent the results of tests by EPA, operators of gas turbines, and
other manufacturers.  Although one might expect that water injection would be
more effective due to the latent heat of vaporization, many investigators
have found only minor differences between water and steam injection.  This is
explained by the superior mixing characteristics of steam and the finite
                                          54
evaporization time of the liquid droplets.
      The industry readily accepts the fact that water and steam injection
are valid techniques for reducing NO  emissions from gas turbines.  Turbodyne
                                    J\
routinely guarantees an NO  emission limit of 75 ppm at 15 percent stack gas
                          J\
oxygen and Turbo Power and Marine currently offers water and steam injection
on an "as required basis".  '     General Electric states that water or steam
injection may be applied to any machine in the product line, that water injection
is currently applied to, some MS5000 and MS7000 series machines, and that  steam  ._
injection has been applied to the MS5000 series turbines for load augmentation
on industrial machines since 1961.    General Electric also states that steam
injection does not appear to affect turbine life and an assessment of the effect
                                      4-24

-------
zuu

180


160

140
5-120
**
k.
O)
a.
m
>^-
H
1 100
E
a.
a.
co"
O
i »°
s
UJ
x
0
z



60

40


20

n










? ?

-1-

i
i
^0.5

















<;









<

<






<



0.3















?










10.52 f
A- i
f^
M
0.9
^




f 4
\





	 , 	 ,
+,
9



p







(






0.43
1
r












0.56



9

















0








j
J0.7






3
<




















LEGEND

^ [ ] Combustor rig test
	 Amount of reduction
0.5, etc. Water/fuel ratio
NOTES
9 The lack of brackets
I G.T. Size notation



<






- (


'1 1
f

,'••







on a
indicates
a field or engine test.

9
r
9


9
i

0.5
f

IDS
O
T
Ii

J0.7 j
T >75
J" t 4" 9
4" i- f"




1
f


FACILITY CODE G2 P VI V2 V3 W Y Z2 FA HA1 U1 U2 VI
T'ABLENO. 8 21 28 30 31 33 35 37 45 47 25 26 28
G.T. SIZE 0.5 2.5 17.2 17.2 17.5 21.3 33 33 52 60.4 13 13 17.2
PUEL TYPE 1


O


9
_i
T


0.31
rt
T
	 L
1
1
x 0.5!
5 t -^
1
t
•i
\*/
1.01
O

V4 HA2
32 48
17.5 61.5
« 	 imuin FIIFI (nisTin flTFt . .^^ . .. NATURAL GAS U
Fi pure  4-13. Summary of NOX emission data from gas turbines using wet control techniques.
                                     4-25

-------
of water Injection on life,  and maintenance Is  not possible to obtain because
of limited time (4000 hours) that their machines have operated with water
                                                                           CO
Injection although the machines have performed  well within this time frame.
As of January 31, 1975, San  Diego Gas and Electric Company operated a gas
                                                                       59
turbine with water injection for 17,951 hours without turbine problems.
      Table 4.2 shows a partial listing of 74 turbine installations for utilities
and process plants, existing or being installed, which use water or steam
Injection to control NO  emissions.   Where available, the hours accumulated
                       A
using water or steam injection are also noted.
4.5.T1.1.1  Water Quality for Injection into the Turbine Combustor - To reduce
corrosion of the gas turbine components, only high purity water must be used
for long-term water injection.  The amount and  quality of water injected into
any turbine will vary from manufacturer to manufacturer and from model to model
produced by a company.  Water treatment requirements will also vary depending
on the fuel burned by the turbine, since manufacturers commonly specify the
combined total quantity of elements in fuel, water, and air which must not be
exceeded.  Table 4.3 summarizes typical water quality specifications for gas
turbines and compares them to boiler feed water requirements.  Obviously the
boiler feed water requirements are more stringent than those for the water
Injected Into gas turbines.   Utilities, therefore, commonly use equivalent water
purification systems.
      Table 4.4 presents a comparison of the consumptive water utilization
for different schemes of generating 60 MW of power.  The steam injection at
15 percent of the gas turbine air flow was used for power augmentation and
represents a water/fuel ratio of about 7.5.  Figure 4.14 shows that a 70 to
90 percent reduction in NO  emissions can be obtained with a 1.0 water/fuel
                          n
                                      4-26

-------
   90
   80
   70
   60
   50
oc
 X
o
LU
O
*  40
   30
   20
   10
               Q NATURAL GAS
               O LIQUID FUEL
   //
  /   /
  /   /
 / /
/ /
                               /    D
                                 000
                         /    O
                      /DO      o
                           /
                       OD
           /    D    /


                                                     o
                                                     o
               0.2
                                                 1.0
                    0.4         0.6         0.8
                         WATER/FUEL RATIO
Figure 4.14. Effectiveness of water/steam injection in reducing NOX emissions.
                                                                    1.2
                                       4-27

-------
                            TABLE 4.2  kWWK TURBINE  INSTALLATIONS WITH MATER OK STEAM INJECTIOH
                                                     FOR NO  CONTROL
«Att*
STEAM
Hinufecturtr
tltstlnghoust10


H
',




(2
General Electric61


j


GE f OH ORDEA62

i «

(2

Turbodyne 63
Turbo Power a Marine M



Westlnghouse 65
General Electric 66

Turbodyne 67
Turbo Power & Marine 68
Uier
Florida Power I Light
Florida Power Corp.

Kansas Power 1 Light
Jacksonville Elect



Kansal Electric Power
1 Southern Calif. Edison
San Diego Gas i Elect
San Diego Gas & Elect
Houston Power & Light
Tucson Gas & Elect
Iowa Public Service
Kansas Power ft Light
City of Jacksonville
Florida Power Corp.
Portland General Elect.
Arizona Power & Light
) Southern Calif. Edison
Ohio Edison
• Houston Lighting i Power
Burbank Public Service
City of Glendale
City of Pasadena
Southern Calif. Edison
Union Carbide
Union Carbide
Exxon
Southern Calif. Edison
City of Pasadena
Location

Enterprise, Fla.

Abilene. Kansas
Jacksonville, Fla.



Osaka, Japan
Oaggett. Calif.
San Diego, Calif.
Naval Training Center
Houston, Texas
Tucson, Arizona
— -

Jacksonville, Fla.
—
Portland, Oregon
—
Lucerne Valley, Calif
—
Houston, Texas
Burbank, Calif.
Glendale, Calif.
Pasadena. Calif.
Goleta, Calif.
Texas City. Texas
Texas City. Texas
Baytown, Texas

Pasadena. Calif.

Ho"." "ofTiir
ID







NA
NA
17
1
1
1
3
3
1
4
6
1
—
1

1
1
2
2
1
1
7
7
1
Accumulated Hours
line's 	 ~li&5"
—
718
6)4
3194
.857
1233
1008
779
NA
NA
I000/un1t
17.951
100
100
500
•3000 Total
—
—
—
—
—
—
not Installed yet
NA
NA
NA
27.3 each
3000 plus(4)
Date
-—
8-31 -75
8-31-75
8-31-75
1
7-31-75
7-31-75
7-31-75
(31

NA
10-31-75
1-19-75
10-31-75
10-31-75
10-31-75
10-31-75
—
_.
—
—
—
— .

Installed 4/74
Installed 10/73
Installed 5/74
1-13-76

15 years - June 1975
48.000 us of
2-15-73 on one turbine
not installed yet
NA
NA
                                                               TOTAL INSTALLATIONS:  58 KIHIMUM, «ATER
                                                                                     16 STEAM
                    Water Injection system  available but not being used.
                    19 gas turbines using water  Injection - total for both facilities
                    NA - not available
                    Unit was overhauled  at  47,000 hrs. of operation with the last 3000 using steam  Injection.
                                                               4-28

-------
                Table 4.369-70'71-72'73
              Water Quality Specifications
                                                    Boiler
                                     Turbine          Feed
Total Dissolved Solids (TDS)
+ Non Dissolved Solids (PPM)         1.0 -  5.0       0.25
Sodium + Potassium (PPM)                0.5^         0.25
Silica (PPM)                           0.02          0.0
Particle Size                          10
Ph                                   7.0 -  8.5    6.5 - 7.0

' '  Turbo Power & Marine  limits  sodium  to 0.1  PPM
                            4-29

-------
                     Table 4.4 74
    Comparison of Consumptive Water Utilization
  for Different Schemes of Generating 60 MW of Power
 Plant Type
    Nu,. i ear
(33% efficiency)

    Fossil
(40% efficiency)

Stag combined cycle
(38% efficiency)

Stag combined cycle with
steam injection at 15%
of turbine air flow

Simple and Regenerative
Cycle Gas Turbine
                  Water Utilization Ibs/kw-hr
                           6.5  -  6.8


                           4.2  -  4.5


                           2.0  -  2.1



                               3.4


                               0
NOTE:  (1) 3.4 - 2.0 = 1.4 Ib/kw-hr used for steam injection
           @ 7.5 water/fuel ratio
       (2)
                      lb> of water/kw-hr at a 1.0 water/fuel ratio
(3)42
v  '  7f~rt> =23.3 times as  much water used by boiler as for a
             turbine at 1.0 water/fuel  ratio
           7~rt>
           u'ia
                          4-30

-------
ratio.   Using the figures from Table 4.4 and assuming a 1.0 water/fuel  ratio,
the consumptive water required for control  of NO  emissions from the turbine
is about 5 percent of the consumptive water use of a comparable steam boiler.
Since the water requirements for control of NO  emissions are small, they
                                              ^
could readily be handled by water transport and storage if water was not avail-
able at a site, thereby increasing siting flexibility.
      Figure 4.15 shows a water purification system as typically used for gas
turbines using water injection.  The figure shows flowrates for a water treat-
ment system which could be used to provide water injection for operating five
28 MW gas turbines 10 hours per day.  The typical system may use a series of
filters followed by a reverse osmosis unit (RO) and a deionization (DI) or
                                                               80
demineralization unit.   These are further described as follows:
                       a)  Reverse Osmosis  (RO)
      Osmosis is defined as the spontaneous passage of a liquid from a dilute
to a more concentrated solution across a semi permeable membrane  -  that 1s, a
membrane permeable to one component of the solution, but not to the other.
      When solutions of different concentrations are separated by a semipermeable
membrane, solvent (water) will permeate to the more concentrated solution.
This flow may be stopped or the direction reversed with sufficient pressure
applied to the more concentrated solution.  The pressure which produces zero
net flow across the membrane is known as the osmotic pressure.  When the pressure
applied is greater than the osmotic pressure, the solvent flows from the more
concentrated solution.  This is reverse osmosis.
      A typical reverse osmosis system can convert 70 - 90 percent of the feed-
water to pure product, depending on application.  Concentrations of more than
2,000 ppm of total dissolved solids (TDS) can be handled.  Inlet pressures of
                                      4-31

-------
                          Figure  4.1576*77'78'79
         Typical  Water Purification  System  for  Five Gas Turbines
         (28 MW each)  Sized to Operate  the  Turbines 10 Hours Per  Day
influent   	
125,000 gal/day
             Pump
                               Reject water equal  to
                               20-30% of influent
Reverse Osmosis
Module (R. 0.)


IU5
infl
20,
= j T.O H time
uent concentr
000 GPD
     effluent from R. 0.
     (90-95% reduction in TDS,
      70-85% of influent)
     Mixed Bed
     Demineralizer
Daily back wash &
rinse to regenerate
ion exchange units
                                          5000 GPD
                                                        Sewer,
                                                        River or
                                                        Evaporization Ponds
                                                      Neutralizing
                                                         Tank
            Storage
                                    95,000 gal/day turbines
                                    TDS  5 PPM
                                    NA + K + Pb + Va   0.5 PPM
                                    Ph 6.5 - 7.5
                                    4-32

-------
400 - 600 psig are normally employed as the driving force.   The product
water will be 85 - 95 percent lower in dissolved ions compared to feedwater
concentrations.  Particulate matter will also be removed.  Some materials
such as iron or high levels of participates can cause problems with reverse osmosis
systems.  To reduce operating costs, many reverse osmosis systems include a
series of filters and pretreatment systems.
                       b)  Deionization  (DI)  or  Demineralization
      Demineralization is a process of removing the mineral salts from water
by ion exchange.  Natural waters, whether obtained from wells, rivers, lakes,
or even collected rain water, contain some dissolved salts, silica, and carbon
dioxide.  The salts are ionized and are present as metallic cations (Ca++, Mg++,
Na+) and anions (ClI SO "~, HCO%).  The silica and carbon dioxide are dissolved,
                       4        J
but ionized to a considerably lesser degree.
      There are two basic steps in DI.  In the primary stage, a hydrogen cation
exchange resin converts the salts to the corresponding acid by exchanging
hydrogen ions (H+) for the metallic cations (Ca++, Mg++, Na+).  After passage
through this acid regenerated resin bed, the water contains a mixture of acids
corresponding in concentrations to the anions originally present in the waters.
These are most commonly carbonic acid (bicarbonates), hydrochloric acid
(chlorides), sulfuric acid (sulfate) and silicic acid (silica).  For example,
H2Z + CaS04  j  CaZ + H2S04.
      The second phase consists of removing these acids by an alkali-regenerated
anion exchange resin.  The most common alkali used is caustic soda (sodium
hydroxide).  The resin may be either of two types:
                                     4-33

-------
      (1)  A "weakly basic" anlon exchange resin which removes  only strong
"mineral acids", s' ch as hydrochloric,  sulfuric and nitric,  etc.,  and permits
"weak" carbonic and silicic adds to pass through.
      (2)  A "strongly basic anion exchange resin which removes "weak" carbonic
and silicic acid as well as "strong" mineral  acids, replacing the  chloride,
sulfa.e, and bicarbonate anions with hydroxyl  (OH") Ions.  When carbon dioxide
and silica removal is desired, a strong base anlon  resin 1s  used.   Where their
removal  is not required, a weak base anlon resin can be used.  When mineral
acids in the cation effluent are high enough to justify the  additional invest-
ment in resin and equipment, a system consisting of a cation unit, a weak base
anion unit, and a strong base anlon unit in series  can be used.
      In practice, RO by itself will not be adequate to achieve the quality of
water required for injection.  RO can,  however, act to significantly reduce the
load on the DI unit, allowing these units to be smaller and  require less chemicals
for regeneration.
      A choice exists in regard to the configuration of the  DI  unit.  This
choice involves the decision whether a/dual bed deionlzer should be used with
separate cation and anion exchange resins or whether a mixed bed system should
be used.  Also, some companies feel .that the dual bed should be used in
combination with the mixed bed system.   The dual bed, followed by a mixed bed,
has a higher capital cost but lower operating costs since less frequent
regeneration is required.
4.5.1.1.2  The Effect of Water or Steam Injection on Emissions  of  Particulates,
CO, and_HC  -  The data reported in the literature is contradictory regarding
the effect of water or steam injection on the emissions of particulates, CO and
HC.  In tests performed by Westinghouse on the model W-251 gas turbine, water
injection resulted in slightly lower HC emissions at base and peak loads, about
                                      4-34

-------
a 75 percent reduction in CO emissions at base load, and a reduction in visible
                                                         81
emissions by 1.5 to 1.75 smokespot numbers (ASTM-D-2156).    Data reported from
tests of the New York Power Pool turbines resulted in the conclusion that the
general effect of water injection on a Turbo Power and Marine FT4 turbine
is to reduce CO, HC, and particulate emissions while the turbine operates at
base load but to increase these emissions when the turbine operates at peak
load.82  Tests performed by General Motors show that the effect of water injection
on CO is a function of the water injection method and that NOX reductions can be
                              83
obtained without CO increases.
      Therefore, it appears that the effect of water or steam injection on
emissions of HC, CO, and particulates is related to the water injection method
and increases in these pollutants can be avoided by proper design of the
injection system.
4.5.1.1.3  The Effect of Hater or Steam Injection on the Conversion of Fuel
Bound Nitrogen to NO  (Organic NQXJ  -  As discussed in Section 3.2.2.4.2,
the exact mechanism for formation of organic NO  is not known, and organic
                                               ^
NO  is only a problem when burning residual oils, some crude oils, or heavy
  /\
distillate fuels which have high nitrogen contents.  Section 3.2.2.4.2 also
shows  that most  light distillate fuels contain less than 0.015 percent by weight
nitrogen, crude  oils generally contain less than 0.2 percent, and residual oils
can go as high as 2 percent.
       Data concerning the effect of water injection on the conversion of fuel
bound  nitrogen is scarce.  General Electric has performed  some very limited
tests  on a combustor rig using  "nitrogen doped" fuels and by using this data
in conjunction with some field test data have extrapolated the results to
predict NO  emissions from an MS-7001C simple cycle turbine burning fuels with
          /\
                                       4-35

-------
                                                                  •   r-      /I  ic
different nitrogen contents at three firing temperatures  as  shown  in Figure a-lb.
The water to fuel ratio has been approximated and added to the figure by assuming
a fuel to air r-tlo of 0.015, 0.018, and 0.021 at firing temperatures of 1600 F,
1800°F, and ?')00°F, respectively.

      The curves presented in Fiqure 4.16 represent total NO  emissions - thermally
                                                            A
formed NO  plus organic NO .  Looking at the curves for the 0,5 percent nitrogen
         X                X
fuel, one can observe that beyond a water to fuel ratio of about 1.2 to
1.3,  the injection of additional water results in increases in emissions of NO  •
                                                                              X
This  would indicate that, beyond certain high rates of water injection, an
increased yield of organic NO  may result.  These water injection rates, however,
                             A
are substantially greater than would commonly be used for NO  control, since a
                                                            A
1.0 water/fuel ratio will provide 70 to 90 percent reduction in thermal NO
                                                                          A
as explained in section 4.5.1.2.  Also, some of the water injection rates shown
in Figure 4.16 would exceed  the  levels considered ab a maximum for an operating
                                                    De-
gas turbine because the flame would be extinguished.

4.5.1.1.4  T_he_^Effect  of  Water  and  Steam  Injection on Gas Turbine Efficiency -  Water
and steam injection for control  of NO  are reported to effect gas turbine
efficiency.  Table 4.3 summarizes the data reported by manufacturers and users
of gas turbines  in response  to  queries by EPA.  The data  reveal that water
injection reduces gas  turbine efficiency by  about one percent for a water  to fuel
ratio of one.  If waste steam is available from a boiler, steam injection  at a
water to fuel ratio of one results  in about  a one percent increase  in gas  turbine
efficiency.  For a combined  cycle turbine using steam from  the boiler portion,
there is an overall reduction in efficiency  of one percent  at a 1.0 steam/fuel
irjection rate.I/
4.5.1.1.5  The Effect  of  Very High Water or  Steam Injection Rates on Mass Emissions

of NO -  Figures 4.17 and 4.18  show calculated NO  reductions as a function of
water and steam injection rates, respectively.   In Figure 4.14, it  can  be
observed that the  effectiveness  of water or  steam injection in reducing NO
                                                                          X
emissions is sharply reduced beyond about a  1.0 water to  fuel ratio.
-/A one  percent decrease in overall turbine efficiency corresponds to about
a 5  percent increase in fuel consumption.       4.35

-------
                          TTFI
      o
      o
      o
      10


       II
  to
  e>
       LJJ

       CC
      cc oc
  r-
  CO

                                                                         .8
                                                                         E"
    CD



    C
    ra
    in
    •!-•

    0}

    C
    o
    o

    c
     O

                                                                         ii- o
                                      4-37

-------










*
p














an
CD
t— >•
O U
LU z
<~3 UJ

— . <_>

S tu
<2 u.
UJ UJ

c/> y4
a •-«
15 co
«c o.
o: ?
UJ
i"**~ VI •
is is
U- IT.
° 5

\— zr
0 0
Lul ^*
U. J—
u. <
LU t—
•sz
LU Ul
=c z:
1— O -i

^f
tr>
• a:
*f Ul
3:

t— a.
.
1 l'
c~
5t
••(V
l_ g
1 £

OJ


a-
™
C d>

g- ^
o a:

—4-X 4->
CT rtj
(U CU
o re
ttl C
~QJ «r—

_i-
O fc
>O f
f>^ ^C
10 3
— "U 4J
rc CO

~.

	 1

S- -
O 1!
c
V. -r-
W JD
L. k.
3 3
4-> t—
U
3 J-
C 01
•5 -^
s ^ E
•—  CO CM VO CO

O O i— CM r—
CO CO CO *t •*•


TO
4J
C
1

•a CM co
^•••^ ^ni-*
O O CM
• * •
CM i— CO
+ 1 •«•





.
o o o p*. co
SVO CO O VO
CM O> O CM
r^ r? 0 00 00
»•- r^ r—







VO
r»» o oo

oo co o






~





O i- r— Or-






 (O U *O
C r— 4J (U >. 0)
rO U «> 4J U 4->

fc. CJ T>
at a>
5 
CO CO CM







CO

CM OO
B •
CM ««•>
+ H-






a CM o
•a- co vo
O CM *T









r"~ r~B
a »
r^<» r^«
-«• -t-











0 r- r-








CT)
OO
O

"C

0 O
 ai

5 3S c/ii

m
Of—
i
00
c?
•*- U-
0
a.



•<
O LU
•r-
nj  3
C O
8 C
 <1>
*~ T3
^^Ul
CM C
>.|


*^^
r"~|










a
2
V.
u
c
"^
» . »
W 3
r— tt
i- C
a> o -~
0 J3
C 4-1
*U EE <^i
t- p 
OJ ^> -C
<*- <•-
Vtl  f—
r— U
•r-  S
4J 10 4->
(O 

^3 (O *O
a» 5 cy»
4^
TO trt ^™*
*3 6 "n
031-
r~* V* M

-------
      The combined result of the negative effect of water Injection on efficiency
(as discussed In 4.5.1.1.4) and the maximum NO  reduction at about a 1.0 water
                                              A
to fuel  ratio 1s an actual increase 1n the mass emission rate of NO  with
                                                                   /\
additional water.  As shown in Figure 4.17, this phenomenon  occurs  at  water
to fuel  ratios of about 1.4 and 1.9, and above.
      If one assumes that waste steam is available for injection, this phenomen
does not occur  -  as shown in Figure 4.18.  If the affect on efficiency of
producing the steam is included, the phemonmenon will occur as in Figure 4.17.
4.5.1.1.6  The Effect of Water or Steam Injection on the Formation of Ice Fog  -
Ice fog occurs only under certain atmospheric conditions and is a serious local
problem in only a small portion of the United States, primarily Alaska.  Ice fog
occurs at temperatures below about -20°F and consists of highly reflective, small
(mean diameter 3 to 7 microns) ice crystals which are nucleated by air-borne
             90
particulates.    It occurs naturally over lakes and rivers.  Man-made ice fog is
caused by exhausts of automobiles and power plants, etc.
      Ice fog severely restricts visibility since the crystals are long-lived.
Once it has formed, it may continue to plague auto and air traffic for extended
periods.  The formation of ice fog is difficult to predict.
      Test data are not available on the operation of gas turbines in artic
areas, so the actual impact of water or steam injection is unknown.  However,
water or steam injection will significantly increase the moisture content of
the exhaust gases.  Some obvious, but unanswerable questions, are:
      (1)  Will the exhaust gases from a gas turbine penetrate an inversion layer
and escape to the upper atmosphere?
      (2)  What are the climatological conditions which will permit formation of
ice fog in the exhaust of gas turbines?
                                   4-39

-------
 g

 x:

Q
O
DC
O
O
 O
 UJ
 cc
 O
 at
                                                        O BASED ON MAXIMUM EFFECTIVENESS
                                                          OF WATER INJECTION
                                                          BASED ON MINIMUM EFFECTIVENESS

                                                          OF WATER INJECTION
                                                 I      I      I             1      1
I
               0.2     0.4    0.6     0.8     1.0     1.2     1.4    1.6     1.8    2.0     2.2     2.4
                                            WATER/FUEL RATIO
       10
            Figure 4.17   Calculated  NO  Reduction  as a Function of Water Injection Rate
                                               4-40

-------
                                                O BASED ON MAXIMUM EFFECTIVENESS
                                                  OF STEAM INJECTION
                                                •BASED ON MINIMUM EFFECTIVENESS
                                                  OF STEAM INJECTION
     0.2     0.4     0.6     0.8     1,0     U    1.4     1.6     1.8     2.0    22    2.4
                                 HATER/FUEL RATIO
Figure 4.18   Calculated  NO  Reduction as  a Function  of Steam Injection Rate
                                      4-41

-------
 4.5.1.2   Dry  Control Techniques



 4.5.1.2.1   Introduction   -  As shown  in  Chapter  3,  the  greatest  concentrations



 of  pollutants from  _,iis turbines  are found  at  the two  extremes of the turbine



 power  range.   CO  and HC emissions  increase rapidly  as load decreases while  NO
                                                                             J\


 emissions increase  rapidly with  load  increases.   Dry  control techniques  for



 reducing  emissions  of NO  from gas turbines are  defined as those techniques
                        X


 which  use operational or  design  modifications rather  than water  or  steam



 injection.  Proper  designs using these dry techniques do not appear to affect



 gas turbine efficiency.   Because of the  large changes in combustor  environment



 between  low power and high power operation of gas turbines and because of



 differences in the  mechanisms of pollutant formation  (as discussed  in Chapter 3),



 techniques which  reduce pollutants at one  end of the  operating range may increase



 pollutants at the other.  This is  illustrated in Figure 4.19 which  shows a



ischematic of  a combustor  for  low power and high  power operation, the factors



 which  affect  formation of pollutants  and some techniques for reducing pollutant



 formation.



       For conventional gas turbine combustors, it has been demonstrated  that



 CO  and NO  emissions can  be  traded off against each other by changes 1n  operating
          /v


 conditions or air flow distribution.  Since CO emissions decrease with  increasing



 load while NO  emissions  increase  with increasing load  as described in  Chapter  3,
              /\


 there  is an obvious trade-off of CO and  NO  emissions as a function of  gas
                                           /\


 turbine load.  Emission data  obtained from tests of many gas turbine engines



 show that the trade-off of CO and  NO  emissions  from  conventional chambers  will
                                    A


 ge  erally fall within the band as  illustrated in Figure 4.20.  The  use  of



 variable geometry to control  average  temperature levels by controlling  combustor



 <- ir distribution  will result  in  a  trade-off of CO and NO emissions by  shifting
                                                         J\


 the emissions performance point  either up  or down the characteristic combustor
                                     4-42

-------
                                                            RESUL
CAUSES
   Low

   Tin
   Pin
   F/A
  High

   Tin
   Pin
   F/A
                                    COMBUSTION INEFFICIENCY
                                       CARBON MONOXIDE
                                       UNBURNED HYDROCARBONS
                                  r QUENCHING
 LOW
                     POOR COMBUSTION  STABILITY
                     POOR FUEL ATOMIZATION &
                       DISTRIBUTION -,
                                                                                   CURE
HIGH POWER TAKEOFF
                                              POLLUTANTS
EXCESS RESIDENCE TIME
HIGH FLAME TEMP
POOR LOCAL FUEL DISTRIBUTION
                                    INCREASE RESIDENCE TIME
                                      REDUCE FLOW VELOCITY
                                      RETARD MIXING
                                    INCREASE EQUIVALENCE RATIO TO 1
                                    IMPROVE FUEL ATOMIZATION &
                                      DISTRIBUTION
REDUCE RESIDENCE TIME
  INCREASE FLOW VELOCITY
  ENHANCE MIXING
REDUCE EQUIVALENCE RATIO
  TO 0.5-0.7
IMPROVE LOCAL FUEL DISTRIBUTION
                                        OXIDES OF NITROGEN
                                           SMOKE
                          91
            Figure 4.19     Factors  Effecting the Pollutants  From Gas  Turbines
                       at Low Power  (Idle)  and High Power Conditions
                                                  4-43

-------
  100
Q)
ft
r-H


O
O
O
r-t
,0
 O
 •H
 K
 W
   10
 Q)
 •d
 o
 g   1
 O
 ,0
 ^
 cU
 O
   0.1
      0.1
                                    CONVENTIONAL COMBUSTORS

                                      O Model  250

                                      n Model  501

                                      A Advanced Turbofan
I    I  I I I I  I I I	I    I  I 1 I  I I I I   	I   I	I  II I  IlL
               1                   10           .         100

  Oxides  of  Nitrogen Emissions, lbs/1000 Ibs fuel
            QO

  Figure  4.20    CO and NO  Performance of Conventional  Gas Turbine Combustors
                       A
                                 4-44

-------
performance line as shown 1n Figure 4.21.  This shift could be obtained by



minor redesign of a given combustor using fixed geometry to obtain a NO  - CO
                                                                       n


trade-off at a given operating point but the use of variable geometry may



then be required to operate the engine over the full load range.



      An undesirable feature of this trade-off is that it merely trades one



pollutant for another.  Reductions of both pollutants (CO and NO ) will require
                                                                A


new combustor designs which either shift the CO - NO  curve to the left as
                                                    A


shown in Figure 4.20 or change the CO - NO  characteristic trade-off curve as
                                          A


shown in Figure 4.22 for a prototype prechamber combustor using variable



geometry.



      Combustor designs utilizing dry control techniques to retard the formation



of thermal NO  usually involve design modifications to influence:
             A


      a.  the reaction flame temperature



      b.  residence time of the gases at temperature



      c.  the amounts of oxygen available for conversion to NO
                                                              A


      d.   atomization  and  vaporization  of the  fuel



      e.  mixing of the fuel and air



The effects of variations in these factors have been discussed 1n Chapter 3



and, therefore, will not be addressed again.  Combustor design modifications



which are applied to accomplish the factors listed above include techniques



such as:



      a.  air staging and redistribution



      b.  fuel vaporization



      c.  fuel staging



      d.  two stage combustion and off stolchiometric combustion



      e.  premixing of the air and fuel prior to introduction to the combustion



chamber



      f.  variable combustor geometry






                                  4-45 .

-------
  5000   .—
 CM

0 1000

X1
in
o
4-''


T5
0)
-P
o
0)
o
i'L.
(0
f.
y,
o
c
o
o
,0
ft
pj
o
                      Variable Geometry


                             CotubustJGr
    100
    10
        ILUJ
                                                                 \
                             5      10                  50       100

              Oxides of Nitrogen Emission PPM Corrected to  1.5^
                                                             50(
        Figure 4.21
                   93
Effect of Variable Geometry Control  on Exhaust  Emissions

    from the T-63 Gas Turbine Engine Combustor
                                        4-46

-------
  5000
  1000
   500
in
(U
&
o
o
o
0<

w  100

o
    50
O
c
c
o
O
    10
— ••»>• Prcchambf
	 — — •- Basic Con
—
—
	
1 1 !
1 5
r Combu
buster



1 1 II
ter
\
\


IDLE^,^.^
1 1 1
*
IDLE
\°
\
A
/ 0\
\ 1 M



«tfi MAX
"^ POWER
V
fo MAX
POWER
1 1 1
10 50 100 <^n<
                 Oxides of Nitrogen  Emissions-PPM Corrected  to 15$ Oo
                    94
         Figure 4.22    CO versus NO  for Rig Tests of the T-63 Combustor and a
               Prototype T-63 Prechafliber Cofflbustor Us i«9 Variable Geometry
                                         4-47

-------
      g.  exhaust gas recirculation



      h.  catalytic combustion



      i.  extern?"  combustion 1n a larger combustion chamber(s) where the



combustion conditions can be more easily controlled than in a conventional gas



turoine combustor



4.5.1.2.2  The application of dry control techniques  -  The dry control techniques



listed above have been applied individually or in combination by gas turbine



manufacturers to reduce NO  emissions with concurrent reductions in CO emissions.
                          A


Many techniques have been researched in combustor rig tests while some have been



applied in varying degrees to production engines.  Much of the design information



on the application of dry techniques to turbines by specific manufacturers is of



a proprietory nature and cannot be discussed.  This section will, therefore,



summarize the data available from the literature and the non-confidential infor-



mation where dry techniques have been applied with some measure of success.



      Where data shown are from combustor rig tests, one must remember that



substantial effort may be required to incorporate such design changes into a



production engine.  The amount of further development work required will depend



to some extent on whether "full size" combustors were used in the rig tests



and on the sophistification of the simulation of actual combustor inlet



conditions and of engine conditions such as the flow configuration and distribution



downstream of the combustor.  Those configurations tested in rigs also, will still



have to demonstrate durability and life  tests under full engine operating



conditions.



      As shown in Figure 4.23, the use of advanced fuel injection methods such as



a-?, blast fuel atomization have been found to reduce NO  emissions by 10 to 15
                                                      A


percent when compared to the more conventional fuel injection methods.  This is



primarily due to the better fuel and air mixing which reduces any localized

-------
      0

      3.
9
L.
bf>
o
f-t

2
          SO
          40
          30
          20
          10
                     •  Fuel:   Kerosene
                                       Current Design





                                       ,—• Alrblast Fuel  Injection
                                                  Rapid Dilution Air

                                                       Introduction
              .95
  Figure  4.23:7a  Reduction  In  the Level of NO  Emissions at Take-off

Power  from an Aircraft Gas  Turbine Using  CombQstor  Design Modifications
                                          4-49

-------
primary zone combustion at stoichiometric conditions.  Figure 4.23 also shows



that rapid introduction of dilution air to reduce the dwell time of the



combustion gases at '.iigh temperatures resulted 1n a 30 percent reduction 1n NO



emissions.



      As discussed in Chapter 3, reduced levels of NO  emissions may be obtained
                                                     A


by operating with much leaner or much richer average primary zone fuel to air



ratior than stoichiometric.  Figure 4.24 shows this effect and shows that the



use of lean primary zone mixtures, in particular, significantly decreases



emissions of NO .
               /\


      Experimental results for tests performed at Westinghouse on a standard



gas turbine combustor using Number 2 oil with successive modifications to reduce



residence time (II), lean the primary zone (increased air to fuel ratio) with



reduced residence time (III), and a very lean primary zone with reduced



residence time (IV), are shown in Figure 4.25.  The curves demonstrate about a



40 percent reduction in NO  emissions for the modified combustor IV as compared
                          A


to the production combustor I.  Research and development tests on one-half



scale combustors using primary zone leaning showed a 10 to 20 percent reduction


                                           98
in NO  for oil and gas fuels, respectively.
     A


      Results of research and development tests performed by Westinghouse on a



one-half  scale combustor without and with simulated exhaust gas recirculation



(CGR) are shown in Figures 4.26 and 4.27 when burning Number 2 oil and natural



gas, respectively.  The figures show that a 38 percent and 30 percent reduction



in NO  emissions was Accomplished when using EGR for oil and gas combustion,
     /\


rr jjectively.  This was accomplished at an EGR rate of 26 percent and the effect



OH the HC and CO emissions was beneficial.



      Solar, under contract to NASA-Lewis Research Center, conducted rig tests



of "can"  type combustors to demonstrate practical combustor technology for the
                                      4-50

-------
         16
     9
     £
     0


     bo
     O
     c-t

     2

     h
      kc
      o
Combustor Inlet Temperature          a 770*K

Comhustor Inlet Air Pressure         * l.OS Atmosphere

Combustor Fuel-Air Ratio  (Overall)    » 0.024

Coobustor Equivalence Ratio (Overall) * 0.35
                      .4         .8         1.2          1.6


                           Average Primary Zone Equivalence Ratio •
                                                   2.0
                                                             2.4
Figure 4.24  96 Effect of Primary  Zone Equivalence Ratio  on NO   Emissions from
                 an Aircraft  Gas Turbine Combustor  at Simulated Take-off  Power
                                              4-51

-------
  220
  200
  180
  160
  140
in
LU
S 120
x
o
z
uj
   80
   60
   40
   20
O W251-AA PRODUCTION COMBUSTOR

O REDUCED RESIDENCE TIME

A LEANED PRIMARY ZONE WITH REDUCED RESIDENCE TIME

O VERY LEAN PRIMARY ZONE WITH REDUCED RESIDENCE
   TIME
                                                                        —O—
    200
400
600             800            1000

COMBUSTOR TEMPERATURE RISE, °F AT
1200
1400
  Figure 4.2597.  NOX emissions versus combustor temperature rise for a production combustor and several
  modified versions of the production combustor.
                                                 4-52

-------
              100
              80
              «
Combustor 6" (Std)
Fuel No. 2Oil
Air Flow 2lbs/sec
Air Inlet Temp. 600°F
Combustor Pressure 70 psia
  	No Recirculation
      <02=Z1.0%I
  	Recirculation
      I02-18 7%)
                   800     1000     1200    1400     1600
                              Combustor Exit Temperature °F
                                   1800
Figure 4.26
               99
Effect of Exhaust  Gas  Recirculation on NO Emissions
       for  a Combustor burning No. 2  oil

-------
            100
             80
             60
          5  40
  Condjtion.s
Combustor 6"  Std
Fuel Natural Gas
Air Flow 2 Ibs/sec
Air Inlet Temp. 575°F
Combustor Pressure 60 psia
Curve  I No Recirculation
Curve I! Recirculation
      (0, - 19.2*1
                 800
                         1000
                                1200     1400     1600
                              Combustor Exit Temperature °f
                                                     1800
                Dry Volumetric Basis
Figure 4.27101  Effect of Exhaust  Gas  Recirculation  on  NO  Emissions
                       for  a  Combustor  Burning  Natural  Gas
                                        4-54

-------
reduction of pollutants from future generation aircraft turbines.   One goal
was to reduce the emissions of NCL at cruise conditions from the present
levels of approximately 18 to 20 gm N02/kg fuel (220 ppm to 247 ppm at 15
percent 02) to a level of 5 gm N02/kg fuel (62 ppm at 15 percent 02) or less.     -/
During these research and development tests, two basic configurations of combustors
were tested - the Vortex Airblast combustor (VAB) and the Jet Induced Circulation
Combustor (JIC).  Both the VAB and the JIC concepts utilized lean primary zones,
prevaporization of the fuel and premixing of the air and fuel.  The basic VAB
combustor is illustrated in Figure 4.28.
       Figures 4.29, 4.30, 4.31, 4.32, and 4.33 illustrate design modifications
which were made to the basic VAB combustor during the test series to improve
combustion stability, improve the fuel and air mixing and reduce emissions
of NO  , HC, and CO.  The data resulting from tests of the basic and modified
     y\
VAB combustors are shown in Figures 4.34A through 4.34E.  The VAB combustor
demonstrated a NO  level of about 1.1 gm N05/kg fuel at 0 percent humidity
                 A                         C
(13.6 ppm at 15 percent 02) with essentially 100 percent combustion efficiency
at simulated cruise combustor conditions of 5 atmospheres inlet pressure and
1040°F inlet temperature when burning Jet A-l fuel.     This represents about
a 94 percent reduction in NO  emissions as compared to the present generation
aircraft turbine (e.g., 18 to 20 gm N02/kg fuel).  It is interesting to note
that the NO  emissions from the VAB combustor did not show dependency on inlet
           A
pressure, as does a conventional combustor.     Figure 4.34B shows the effects
of fuel on NO  emissions, and Figures 4.34D and E show the effects of combustor
             A
inlet temperature on NO  and CO emissions, respectively.
                       A
       The basic JIC combustor is shown in Figure 4.35 with fuel injection and
combustor design modifications shown in Figure 4.36.  Data results are presented
in Figures 4.37A through E.  As can be observed, the VAB combustor concept
-/ A conversion:  ppm @ 15% 02 = 12.34 X gm N02/kg fuel was used for these
figures.
                                       4-55

-------
            ,SWIRL VANES (24)
            '       _CONVECTIVE
                   \TCOOLINGAIR
       .AIR BLAST FUEL
        INJECTOR TUBES (24)
                                         0.133M
                                         (S.2S IN.I
                                          D1A.
              103
Figure 4.28      Details of the  Basic  Vortex Airblast (VAB) Combustor
                                    4-56

-------
                                IRL VANES (24)
              INITIAL FUEL
              TUBE PENETRATION
                                DRIP FENCE
                                        FUEL

                                        INJECTOKS
             104

Figure 4.29      VAB Combustor - Fuel Injection Modification
                              4-57

-------
                       0.133m
                      (5.25 IN.) DIA    ORIGINAL REACTION
                                  ZONE DIA
                  INCREASED REACTION
                  ZONE DIA          0.20m
                               (7.88 IN.) DIA
              1 nc
Figure 4.30      VAB Combustor -  Reaction  Zone  Diameter  Increase
                                    4-58

-------
                                    EXTENDED
                                 SWIRLER CHANNEL
                                     EXTENT OF
                                     ORIGINAL
                                     CENTER BODY
                                  0.051 m(2.0IN.)
Figure  4.31      VAB Combustor - Swirler Throat Length Increase
                            4-59

-------
Figure 4.32     VAB Combustion -  Reaction Zone  Length Increase
                        4-60

-------
                                  fUEL RAKE (24)
                                          FUEL
                                          INJECTOR
Figure 4.33108  VAB Fuel  Injection Modification
                                4-fil

-------
1300  1400 1500 1600 1700  1800  1900  20002100
       COMOUSTOR TEMP. RISE(DCG.  F)
      700
               800      900      1000
              COMBUSTOR TEMP; RISE - (DEC. K)
                                          1100
                                                            —
                                                            5
                                                               5.0


                                                               4.0



                                                               J.O



                                                               2.0



                                                               1.0


                                                               c  0
                                                                       TIN-833' K (1040V)
                                                                       PIN-204.8KP»<15»ilj)

                                                                                   VAB COMBUSTOR


                                                                              • 2  DIESEL
                                                                                                 3.93 CM
                                                                                                 NOz/KGFUEL
                                                                           GASOLINE
                                                                                                     1.18 CM
                                                                                                     N02/KC. FUEL
                                                                    1100   1200
                                                                              1300   1400   1500  1600   1700  1800
                                                                               TEMP. RISE-4F. OEG.)
                                                                   600   650
                                                                            700750   800   850
                                                                              TEMP. RISE-4K. DEC.)
                                                                                                      900   950 1000
  Fig.  4.34A
             VAB corcbustor  NO  test results;
                                                       Fig. 4.34B  VAB combustor test results - effect  of
                                                                    fuel type
 1.2



 i.o


-------
                                   AIR OLAST
                                   FUEL INJtCTION
                                   TUBE (4)
                        <4)AIR/rUEL
                        MIXING PORTS
          CONVECTIVE COOLING
      TORCH IGNITER
      ENTRV POINT
                               0.533M.
                               (21 IN.)
Figure 4.35
              112
Basic Jet
Details
Induced Circulation  (JIC) Combustion
                               4-63

-------
                                   SINGLE POINT
                                   FUEL INJECTION
                                     MIXING TUBE
                                     ENTRY
                                    MULTI POINT
                                    FUEL INJECTION
Figure     A     JIC Combustor  Fuel  Injector
                 Modification
INCREASED REACTION
ZONE VOLUME
                                                                                0.133 M
                                                                                (5.25 IM.)DIA.
                                                                            ORIGINAL  ,
                                                                           RCACTION
                                                                           ZONE VOLUME
                                                                          0.191 M
                                                                          (7.5 IN.)
                                                                          DIA.
                                                         Figure   B    JIC Combustor  Reaction  Zone
                                                                        Volume  Increase
                                             ORIGINAL MIXING
                                             TUBE PENETRATION
                                             FLUSH MIXING
                                             TUBE MODIFICATION
                          Figure    C   JIC  Combustor Mixing  Tube Modifications
                                      113
                        Figure 4.36     JIC Combustor Modifications
                                                   4-64

-------
U.  5.0
*J3
X

o  4,0

a
y  3.0

X
o  2.0

V
I  1.0
jic ruMnusroR
 fun.  .111-A i
 NOMINAl (.OMMUSTOR INI ET CONDITIONS:
       im 'K (1040 T)
           KPAI2S PSIA)
INITIAL COfiriGtmATION-
          I
                        -MIXING TUBES FLUSH
                         WITH REACTION ZONE
                               I    I 	I
    1250 1300 1350 1400  1150 1500  1550 1600  1650 1700
              COMBUSTOR TEMPERATURE RISE -(DEC fl
     -J	1	1	I	I  '       1
     700      750     800 •     850      900      950
              COMBUSTOR TEMPERATURE RISE -(DEC. K)

     Figure A  JIC combustor NOX test remits
               ^750805850"
                TEMP. RISE-IK. DEC.)
                                                                                                900
                                                                                                       95C
                                                       Figure  B   JIC combustor test  results -- effect of
                                                                  fuel  type
                       JIC COMBUSTOR
                       FUEL; JET-A1
                       T IN ="756 DEC. K
                            (900 DEC. F)
                                       £,  PIN- 204.B KPA
                                                (15 PSIG)
     0   2200   2300    2400    2500    2600   2700
              COMBUSTOR OUTLET TEMPERATURE -tT)
                                           2800
          150T>	1600         1700
              COMBUSTOR OUTLET TEMPERATURE -OO
                                        TBT
  Figure  C  JIC combustor  test  results — effect of
             combustor Inlet  pressure
                                  45
                               530
                               u
                               §20
                        I 15
                        VI
                        i 10
                        u
                        s  5
                           0
                                                           4.5
o
" 2.5
x
UJ M
o 2
                                                                2  0.5
                                                                 JIC COMBUSTOR             to " 7
                                                                 FUEL: JET-A1                (900 TL
                                                                 COMB. PRESS. - 204.8 KPA (15 PSIG)
                                                                          'I*
                                                                     • 4J1'K
                                                                      (315T)
                                                                2200   2300    2400    2500   2600   2700   2800
                                                                     COMBUSTOR OUTLET TEMPERATURE -(DEC. F)
1500         1600  .        1700
  COMBUSTOR OUTLET TEMPERATURE •(DEC. K)
                                                 1SOO
                                                        Figure D  JIC combustor  NO   test results-- effect of
                                                                  combustor Inlet temperature
                                                     JIC COMBUSTOR
                                                     FUEL: JET-A1
                                                     COMB. PRESS. • 204.B KPA (15 PSIG)
                                      Tin-831
                                         (1035 T)
                                      2200   2300   2400   2500    2600    2700    2800
                                           COMBUSTOR OUTLET TEMPERATURE - (T)
                                     	   i       	I	I	I
                                         1SOO          1600          1700
                                           COMBUSTOR OUTLET TEMPERATURE -CK)
                                                                                1800
                                Figure  E   JIC combustor CO  test  results—effect of  r
                                           combustor Inlet temperature
                    Figure  4.37114A,  B,  C,  D, E

                        Results from Rig Tests  of  the JIC  Combustor
                                                        4-65

-------
exhibited greater N02 reduction potential than the .)IC concept although both



concepts showed very promising results.



      As a result ' * tests on the OIC and VAB combustors, Solar concluded that:



(1)  lean reaction, pre-mixed systems are capable of demonstrating NO  levels
                                                                     t\


considerably lower th<_n those of current technology designs; (2) NO  levels
                                                                   X


as low as 1.0 gm NO^/kg fuel (12.34 ppm @ 15 percent Op) at cruise condition



were demonstrated with very low HC and CO levels; and (3) further development of



these concepts is necessary to extend combustor operation over the full load



range anticipated by future aircraft engines; possibly by fuel staging or


                  115
variable geometry.



      Advanced combustor designs using features to provide favorable primary



combustion zone fuel to air ratios at idle as well as lean and uniform primary



zone mixtures at high engine power operating conditions are currently under



development by General Electric Company and Pratt and Whitney Aircraft under



the NASA experimental clean combustor program.  The objective of these programs



is to develop technology for use in existing and advanced aircraft turbines



to meet the 1979 standards for emissions of CO, HC, NO  , and smoke as defined
                                                      /\


by EPA for aircraft.  The program objective is to meet  the 1979 standards without



the use of water injection to control NO  emissions.
                                        A


      For Phase I of their contract under the NASA clean combustor program,



General Electric performed rig tests of four basic full scale combustor design



concepts and several design modifications of the basic  four.  The four



configurations are:



      1)  A single annular lean dome combustor



      2)  A double annular lean dome combustor



      3)  A radial/axial staged combustor



      4)  The NASA swirl-can combustor
                                    4-66.

-------
In addition, the standard CF6-50 combustor was tested to obtain baseline
data.  This basic combustor configuration 1s shown 1n Figures 4.38 and 4.39.
      The lean dome single annular combustor 1s shown 1n Figure 4.40.  The
lean dome single annular configuration is basically a CF6-50 combustor modified
to increase the air flow through the primary reaction zone, thereby decreasing
the primary zone equivalence ratio.  Variable geometry would be needed with
this configuration to provide stable combustion throughout the operating load
range.
      Schematics of the lean dome double annular combustor are shown in
Figures 4.41 and 4.42.  In this configuration, both annuli are fueled at high
power conditions with only the outer annul us fueled at low power conditions.
This two stage combustion could, therefore, eliminate the need for variable
geometry as required for the lean dome single annular configuration.  Again,
the primary zone air flow of the double annular combustor was increased to
reduce the equivalence ratio.
      The radial/axial staged combustor configuration is shown in Figures 4.43
and 4.44.  Both radial and axial fuel staging provisions are incorporated.
The pilot stage (as shown in Figure 4.43) is of conventional design and was
sized for low power operation.  Both stages operate at high power conditions
and some degree of fuel and air premixing occurs in the main stage combustor.
Lean combustion in the primary zone was also a design objective in this
configuration.
      The lean dome double annular combustor and the radial/axial combustor
both used airblast fuel injection wherein a portion of the compressor discharge
air is utilized to atomize the fuel and provide better fuel and air mixing.
The use of airblast fuel injection results in reduced NO  emissions when compared
                                                        A
to a conventional high pressure fuel nozzle as shown in Figure 4.23.
                                     4-67

-------
                                       s-
                                       o
                                       -M
                                       Ol
                                       c
                                       o
                                       un
                                        t
                                       o
                                       o
                                       3
                                       •o
                                       O

                                       a_
                                       oo
                                       CO
                                        a>

                                        3
4-68

-------
                                                                   :i
Figure 4.39^ ?.  Production CF6-50 combustor assembly.
                        4-69

-------
                                   J»
                                   I
                                    M
                                    0
                                    0
                                    &
                                   co

                                   *~o
                                    I
                                    O)
4-70

-------
                                                           
-------
                                        3
                                        .0
                                        e
                                        i
                                         3

                                         C
                                         0)
                                        r-»

                                         3
                                         e
                                         a
                                         OJ
                                         Vl
                                         v
                                         C
                                       o
                                       CM
                                         CM
                                         S-
                                         3
                                         cr>

                                         ul
4-72

-------
                                                           S-
                                                           o
                                                           3
                                                           -Q
                                                           TJ
                                                           «J
                                                           O)
                                                           
                                                           I/)
                                                          T3
                                                           it)
                                                          DC
                                                           QJ
                                                           O)


                                                           I
                                                           O)

                                                           o>
                                                          CD
                                                        CVJ
                                                          CO
                                                            •
                                                          ««d-

                                                           2!
                                                           a
4-73

-------
            ] 99
Figure 4.44     Radial/Axial Staged Combustor for CF6-50 Engine.
                               4-74

-------
      A schematic of the baseline swirl-can combustor design is shown in
Figure 4.45.  This concept consists of a modular array of carburatlng swirl
cans, each with an axial air swirler and a stabilizing plate.  Each module
is designed to premix the fuel with the air in the carburetor, swirl the
fuel-air mixture, stabilize combustion in the swirl can wake and provide
mixing areas between the bypass air through the swirl can array and the hot
gases in the wake of the swirl can modules.  With this configuration, fuel
staging was investigated by controlling fuel flows to the various modules.
      NO  emissions versus fuel-to-air ratio for the CF6-50 production combustor
        ^\
and the best configuration of each major design approach (discussed above)
are shown in Figure 4.46.  The double annular and radial/axial configurations
obviously resulted in the greatest reduction of NO  emissions.  A summary of
                                                  X
emissions from the double annular and radial/axial "best configurations" is
shown in Table 4.6.  The table shows that NO  reductions, as compared to the
                                            A
production CF6-50 turbine, at standard day take-off conditions, were 51.4 and
60 percent for the double annular and radial/axial configurations, respectively.
The reductions in CO and HC emissions at idle conditions are also shown in
Table 4.6.
      For Phase I of their contract under the NASA clean combustor program,
Pratt and Whitney Aircraft Division performed rig tests of three basic full
scale combustor design concepts and several design modifications of the basic
three.  The three configurations are:
      (1)  Swirl-Can combustor concept
      (2)  Staged premix combustor concept
      (3)  Swirl Vorbix combustor concept
Baseline data from earlier testing of the production JT9I>7 annular combustor
were used for comparison purposes.

                                  4-75

-------
                                                        c
                                                        CD
                                                        t-
                                                        o
                                                        o
                                                        o
                                                        3
                                                        -a
                                                         i
                                                        c
                                                         I
                                                        r—
                                                        S-

                                                        '5


                                                        OJ
                                                        c

                                                        r—
                                                        OJ

                                                        (T3
                                                        00
                                                      00
                                                      C\J
                                                        LO
                                                         (U
4-76

-------

0)
p
O
SC

O
+->

•o
0>
+->
0)
r-4

a
X
w
0)
bo
A:

bC
X
0)
c
0
in
•H
e
                                  Production Combustor

                                       (II-l)
                 II-5

          Single Annular  Lean  Dome
                                              1-16 Swirl-Can
                                                    11-16 Double

                                                      Annular
                                                       11-12,15

                                                       Radial/Axial
                                                      ECCP Goal
                                                  Nominal  Takeoff

                                                  Fuel-Air Ratio
     0
0.008
0.012      0.016      0.020      0.024

              Metered Fuel-Air Ratio
0.028
                                                                       0.032
               124
    Figure 4.46     NOX Emissions  Levels,  Best  Configuration of Each Major
                    Design  Approach
    NOTE:   At a  fuel-to-air  ratio  of  0.20, NOX  9 15% 02» 12.3  (Emission Index)
                                   4-77

-------
  u
  3
 X>

 5
  C
  O)
  O  CO
  s-  c
  a> T-
 Q-  en
     c
 X) LU
  C
  re  c
     o
  tO -r-
  C 4->
  o  o
 •r-  3
    C
  CO  CO
  3  t-
  re  s-
  -C  3
  x o

  w  e
  i*-  o
  o  i-
     14-

  ^
  re
  co
LT)
CM
   co

  XI














CO
3
U_

cn

^v^
Dl

*
i
u
•5
C

C
0
•r-
E

















^^
c
0
•M T— CO
C 4J C
01 U -r-
t. 3 cr
5- X> C
3 O UJ
!_> I.
Q.
.J r—
0) O
D cn











s_
O fO
5 "3
3 C
O C









re
>r»
X "O

CM 10 ir n




 CM i —
+j r-. •* to m
—
CO
o
_
D
_ ^
CM CM
40 «3- ^"^ ^~hv
CO CM i—

5

j
•o
CO
oi ro CM CM
-M CM CO 00 O
^1 O^ t/^ v£3 VO
1)
o
S-
co
CL.


^~» ^— >.
r~— r-"
CM co •* ^r
CM i— i—

{ ?






II
lf_
o
CO
CO CO ^
i— i— <4- >
>, X, J«J , S-
i- i- (O rtJ
(O fO D T3
TJ XJ C
C C •»->  4J 3: to
co co

+•> >4-> rO rO
(O re
X X
o o o o
3: o z z






z.
00

LU
<^
co

•»
X
CO
x»
C
1— ^
•g
E
LT)
r-™
^



V











LO
^_ •

N

















0>
^y'


co

.^
fl3
CO
a.












XJ
CO
o
CO
o
Jh..
a.

re

C
•1—

^
CO
S—
JC
a
•r—
+3
sp_
a.
1
IH

^-i.
CO
H3
M—
JZ
4^
•r—
3E
r— LO r—
CO 00 CO
3* * 2*
a> o> co
r— Cft i—
c»«- oo
a_ o Q.
CO r— CO
C CO C
•r- > -r-
O) CO CJ>
C i— C
CO CO
>.
co o cu
3 C 3
S- CO S-
4^ «r- +J
O
O -i- O
4-> <4- •)->
t^_
X3 CO XJ
CO CO
4J C 4J
re o re
O 4-> O
CX (/I CX
fO 23 'O
S- XI S-
+-> E •*-»
X 0 X
LU O LU
^
^•^ ^""^
i— CM
* — 1 * — '
                                               4-78

-------
      All tests were performed using Jet-A fuel.   Figure 4.47 is a schematic
of the swirl can combustor concept.  This combustor consists of a three row
array simulating 120 swirl modules for the entire annulus.  Only the outer
module row is fueled for low power operation.  This combustor configuration
is similar to that described earlier for the GE swirl  can combustor.
      A schematic of the staged premix combustor concept is shown in Figure 4.48.
This combustor consists of a primary or low power burner and a secondary high
power burner with each burner having its own fuel injection system and pre-
mixing passage.  Both burners operate during the high  power conditions.  Since
HC and CO are the main pollutants of concern at low power, the pilot burner was
designed to operate at an equivalence ratio of 1.0 during low power operation.
At high power operation, both burners were designed to operate at an equivalence
ratio of approximately 0.7.
      Figure 4.49 shows a schematic of the swirl  vorbix combustor concept.
This combustor employs two burning zones.  The pilot combustion zone only is
fueled during low power conditions with both burners used for high power
conditions.  Fuel for the main combustion chamber is introduced and vaporized
at the exit of the pilot zone.  The objective of the swirl vorbix combustor
is to provide a relatively long combustion residence time at low power settings
to minimize CO and HC emissions and to provide rapid burning and quenching of
the combustion reaction at high power levels to minimize NO  formation.  The
                                                           X
combustor features a combination of vortex burning and mixing using swirlers
as shown in Figure 4.49A and B.
      Table 4.7 summarizes the test results for the best configurations of each
combustor concept at sea-level take-off operations.  NO  reductions ranged from
                                                       y\
34.6 to 60.6 percent when compared to the NO  emissions from the production
                                     4-79

-------
                        STRUT
                                            FLAME STABILIZERS
                                           AND BLOCKAGE ARRAY
                    TYPICAL SWIRLER
                                   CARBURETOR
                                   ARRAY
                                   (40 PER ROW (ANNULAR BASIS)
CARBURETOR  SWIRLER   FLAME
                      STABILIZER
          FUEL INJECTOR
                                  MODULE
                                  COMPONENTS
             Figure 4.47^6 Swirl-can combustor concept.
                             4-80

-------
                           PRESSURE
                           ATOMIZING NOZZLE
0
                    DILUTION AIM
                                                  MAIN BURNER  •
 40 PILOT FUEL
 INJECTORS
PRESSURE
ATOMIZING
'NOZZLE
                                                        .DILUTION AIR
              40 MAIN FUEL
              INJECTORS
MAIN PREMIXING PASSAGE
                           PILOT
                           PREMIXING PASSAGE
                                            40 PI LOT FUEL
                                            INJECTORS
                      CONVOLUTED PILOT
                      PREMIXING PASSAGE
                                                          40 MAIN FUEL
                                                          INJECTORS
                                                              PILOT FLAME HOLDER
          PREMIXING PASSAGE FRONT END VIEW
                                             MAIN        E^ O  TRANSPIRATION 'O i
                                             PLAMEHOLDER *\O O  O COOLED SECTION >
                                         PILOT AND MAIN BURNER FLAMEHOLDERS
                                            127
                               Figure  4.48      Staged Premix Combustor Concept
                                                   4-81

-------
PILOT INFLOW SWIRLER
       AIR FLOW
2 CELL
HECIRCULATION      (g
PATTERN     FINWALL^
           /TANGENTIAL
          J^INFLOW
           FUEL
           NOZZLE
/
                                                           60 INNER MAIN
                                                           SWIRLERS
    CYLINDRICAL
    PORTION OF PILOT         ^
             PILOT COMBUSTION
  PI LOT ZONE
                                                                 MAIN
                                                        oiLurioNi*-COMBLIST|ON
                                                          AIR    ZONE
              STRUT   /
               (^   60 MAIN
                   INJECTORS
              MAIN
              SWIRLER
                      FRONT VIEW

                    Figure 4.49 A. Schematic of the Swirl Vorbix Combustor.
                                                             MAIN BURNER
                                                           OUTERSWIRLERS
                        MAIN BURNER INNER SWIRLERS
Figure 4.49 B.  View looking upstream toward headplate showing orientation of main burner
inner and outer swirlers relative to the pilot fuel injector nozzle.

                   Figure 4.491^. The Swirl Vorbix Combustor Concept.
                                          4-82

-------
                                         TABLE 4.7

                    SUMMARY FROM TEST RESULTS FOR BEST CONFIGURATIONS  OF
                       EACH COMBUSTOR CONCEPT AT SIA-LEVEL  TAKE-OFF
                                        iCONDITION129
 Combustor Concept
 and Configuration
Oxides of Nitrogen
Emission Index
Corrected to Engine
Design Take-off
Conditions (g/kg fuel)
Percent Reduction
  from JT9D-7
Production Combustor
Combustion
Ef f i ci ency
(Percent)
SAE Smoke
 Number
 Goals
 JT9D-7 Combustor

 Swirl Can
 Configuration N9

 Staged Premix
 Configuration P3

  virl Vorbix
"^-configuration S10
         10.0
         31.5

         13.6
         20.6
         12.4
      56.8
      34.6
      60.6
    99.0
    99.99

    99.5
    99.4
    99.7
    15
    10

     1
    14
 NOTES:  Oxides of Nitrogen data corrected to engine conditions with combustor
         inlet pressure of 21.7 atm, combustor inlet temperature of 768.9 K,
         and combustor inlet airflow rate of 92.9 kg/s.

         Combustion efficiency (based on gas sample) and SAE Smoke Number
         are recorded data at test rig conditions with inlet pressure of 6.8  atm,
         inlet temperature of 768.9 K, and inlet airflow rate of 6.88 kg/s.
                                               4-83

-------
JT9D-7 combustor with the swirl  vorbix configuration exhibiting the greatest



potential for NO  reduction.  The NOX data shown in the table were extrapolated



by Pratt and Whitney t.j engine conditions.



4.5.1.2.3  Catalyticajly supported thermal combustion  -  Catalytical'ly supported



therma"1 combustion is an emerging and promising concept for the reduction of



NO , CO and HC emissions from gas turbines.   Catalytically supported thermal
  A

                                                                        130 131
combusl-'cn is believed to involve both catalytic and thermal  combustion.    '



A schematic of a catalytic combustor is shown in Figure 4.50.  In the catalytic



combustor, fuel and air are premixed prior to introduction to the catalyst



core where the reaction begins and combustion is completed at reaction temperatures



below 2800°F; well below the threshold for thermal  nitrogen fixation.133*134'135'136



For comparison, the combustor axial temperature gradients for conventional, lean



premix and catalytic combustor concepts are shown schematically in Figure 4-51.



      In 1973, Westinghouse, in a joint effort with Engelhard Industries,



performed a series of tests to determine the feasibility of using combustion



catalysts in stationary gas turbines.  Table 4-8 summarizes the results of



testing a conventional combustor and a catalytic combustor in the same test rig



for combustion of coal gas and Number 2 distillate fuels.  The catalytic combustor,



as compared to the conventional  combustor, reduced NO  emissions by more than
                                                     A


98 and 89 percent when burning distillate oil and coal gas, respectively, with



concurrent reductions in CO and HC emissions.  It should be noted that these



reductions were accomplished with the catalytic combustor operating at a higher



outU . temperature than the conventional combustor.  Table 4-9 shows results



of additional testing in 1974 and compares emissions from a lean burn premix



com1 ustor to those of the catalytic combustor.  The catalytic combustor again



had very low emissions which were of about the same order of magnitude as the



emissions from the premix configuration.
                                       4-84

-------
CATALYST CORE
COMBUSTION AIR IN-
             /-
    LI
        FUEL

         IN


NO BURNING HERE


      NO COOLING AIR
     FUEL
 PRESENTATION
    \  i j \  \
                           REACTION
                           —••ZONE—
                           "' "'_.'.('.' ~
                                            I
                             DILUTION AIR IN
   TO
                                                  1
TURBINE

	hH	
                                       PEAK
                                       TEMPERATUR
              CONTROLLED
              COPVIBUSTION HERE
                                                  TEMPERATURE
                                                  PROFILE TO
                                                  TURBINE
        ,132
FIGURE 4-50.    Catalytic Combustor Design Concept
              4-85

-------
UJ
K.
ID
CL      4000

£ 2000-
LU
c:
t-
en
       3000
   1500-
   ,ooo
       2000 •
o
o

UJ
o
<
c:
UJ
  500-1000-
    0-0
                   •LEAN PRE-MIX COMBUSTION
                                             CONVENTIONAL COMBUSTION
                            SIGNIFICANT
                           /  NOx
                            FORMATION
              COMBUSTOR
              INLET
CATALYTICALLY  SUPPORTED
      COMBUSTION
                                AXIAL DISTANCE-
                                                          COMBUSTOR
                                                              OUTLET I

            FIGURE 4-51.
                      137
                          Temperature History for Conventional Combustors,
                          Lean Pre-Mix Combustors, and Cata^ytically
                          Supported Combustion
                                   4-86

-------






-o
s_
r-

a>
o>
c
UJ

CO
3
o

en
c:
to c
CO O
+j"
0)  r—
tO
O LU
• i r
*+•• *—
O
CO •<->
•(-> CO

3 O
co E
CD O
ce o

4-> O
CO •>-
CO •«->
t— >>
r—
co re
en to
CO
ro
r~*
00
1
^f

CO
to






xz:
0 0-
n: o.
s:
o±
O Q-

O Qu
Z (X


m
C/)

L^J ^— •«,
a: to

ji^j
fV v_>
P-
co

CO  O
f^









LU
£







0
1—
oo uj
ra Q.
CQ >^
8
00

CO

O
CO



,— ^
VO

en
CO
CM





CM





^•^
cvj
CO
en


o
o
CO







UJ
h-
_i
i
h-H
1— 1
O

CM

=1t=


0
t— 1
H-

LU
8
§

o
CO
r—





^— .
CO
•
CM
in
CO




f-M
r_
in





*«*~»»
o
en
en


CM

t*^i
r™









s

— J
^t
o
0








1 «*

O Q


V V



,— ^ ^-v
CO **

en co
CO O
CO CO




LO O
co «*
LO *t




, — , f-~.
en vo
CO LO
i— r—


CM CM
CO •—
O •—
CM CM






LU
I—
3
P CO
CO  0
3 -J^
0) (0
JZ 3
"(0
E >
O UJ
M- i-
co •»->
+J CO
r— 3
3 ^Q
CO £~
CO O
0£ O
•M O
CO -r-
0) •!->
K- >>
r—
**• (O
r*"*» 4^
en «
r- 0
en
CO
en
i
^^

CO
J3
»2






xz:
0 0_
3C 0.

si

O 0.
z o.


co
CO
Jjj ^"^
C^ fO
o. o.
^^
Q£ V>^
g
p>v
to
=?

3E* H— i
SCO
0.




.
fy
1 1 1 ^j
H-O
^^llll'
t-
LU

H~ ^^
=> o
0









UJ







o
K-
tO LU

£9 2r
S
-

o
CO
LO



^_^
»j-
•
LO
CM
CO
^N—**



CM
•
!*




^.^
en
00
r—


CO
^**
r"~
CM






LU
r™*
_i
_j
CO
1— 1
o

CM

=«:



1—
CO

_J
5
-

vo
CM
r—



^^
f«».
•
CM
O
CO
s^x



en

CO




^^
vo
CM



O
CM
CM
CM









CO
S

j
*^c
o
o








00

o
CM
VO



*~»
0
•
CM
CM
CO
%IM^



f*!^.
•
CO




. — ,
vo
vo
o
p—


o
in
en
r—






LU
h-
3
P
CO
1— 1
a

CM

stts





X
i— i
LU
OH
Q.
4-87,

-------
      Similar low emission levels from catalytic combustion have been verified
                                             on1'
                                             142
and reported in tests by Detroit Diesel  Allison   , NASA Lewis Research Center
and the Air Force A' .*o Propulsion Laboratory.
      Table 4-10 presents emissions data for catalytically supported thermal
combustion when various fuels are burned.  Again, the emission levels are
extremely low except for the "doped" fuels with high nitrogen content (i.e.,
propane with 0.17 weight percent nitrogen as ammonia and Number 2 diesel oil
with 0.94 weight percent nitrogen as Pyridine).  For the "doped" fuels, the
                                                                              144
conversion of fuel bound nitrogen to NO  was in the range of 70 to 90 percent.
                                       X
Engelhard presently has research studies underway to use improved catalysts to
                                          145
reduce conversion of fuel nitrogen to NO .
                                        A
      Although the results show that catalytically supported thermal combustion
is indeed a feasible concept for reduction of emissions from gas turbines,
there are still some areas where additional research is required before a
production gas turbine using catalytic combustion is a reality.  Some of these
are:
      a.  The start-up problem  -  The catalytic combustor must reach a certain
operating temperature, depending on fuel type, for ignition and continued
combustion.
      b.  Variable  geometry will be required to operate the catalytic system
over the full operating  range of a gas turbine.
      c.  The life  of the catalyst and performance over this time period must
be determined.  Engelhard reports that catalysts in  rig tests  have demonstrated
1C'3 hours of operation  without degradation  in performance.     Many stationary
gas turbines operate 8000 hours per year.
      d.  Maintainability and reliability must be demonstrated  in a  full turbine.
                                         4-88

-------








•o
i.
re
-C
r—
O)
CD
c
LU

01
-C
4->
^

•r;

CO
r~
01

U.

U
•_£
^
CO

•o -P
C CO
re >>
i— re
re 4-*
•r- re
4j r—
Pj r— •
O 1
*-* *5
L. 0
0
**"" 3c
re o
•P i —
re <:
0 0
CO

"T

O
r—
1
^"

O)
r~~
^t
re
h—





















,»-^t
>
E
Q.
O.

CO
C
o
•r-
CO
CO
•r-
c
LlJ









•^
.O.
• c
0)
E
0}
J-
3
CO
re
CO
c

CO
CO
•r*
E

s_
o
«*-
co
c
o
•r-
4-*
• pBB
•o
c
o
o




















CO
i-
W
O)
a:
;
O
p^




O
o





y"^
c
o o
=> CO
re




L.
*|"" X"*"!
<: o E
^•i- Ol
•TJ •_* "_•
3 cc. a
U- - — •




x-^s.
in >»'— >
s 	 4-*—
O) -i-l
0 0 •
Wrt t
v^ •»
CO (U **— *
>




^_>
o
o
(U
S-

0) 5
i— re
C i-
i— i a»
a.
E
(U
















,_
a>
3
U_


, 	 , ^-^
CM i—
*— • ^™*

< r— r—






00 O
^* ^"™






) O CO










•r-
a <•
CM CL.
^ "^

*-^ *-^
r- i—
****








CM in
CO CM






V£> CM









f"^ CO
CM r-
O CO
0 0







O O
0 0
O O
O O
«a- o










0 0
O CM
«3- CM















^^
CO
N^X
U
•i- CO
4J re
(U CD
* p p""
4-5 r—
4J c re
,q
O CO O

^^
r—
*™^








O







m









CM
CO
o
o







o
o
o
o
'









o
in
CM










re
0) -r-
c c


Q. 4->
2 co
r*~ re
re ^^
•r- I— C
0 . 0)
i- O Ol
o o

O-'r^-'r^
CS x z

^^
t—
****

o

^^*




CO







CO









in
CM
0
o







0
o
o
CO
r""









s
CO









0)
c
•r-
•o

^o ^^
i— • a.
•i- 4J
O X co

l""~ ^^
0) O> C
CO • 
CO
CO
ai
i.
Q.
4-> X
CO O)

r*— ^

4J 'c
re o
U E
a> o

CO
<*- -r-
o

£ ^* ^
rein re
0) 1
t- CM en
4J Z C
t/) *r~
C •» "O
^C If5 ^0
Or- O
•O 1 r-
CM
10 3= S-
O> O
x: ••« 4J
o en co
C 1 3
T- CM JQ
o e
CMCJ 0
o i— -a
L. I -i-
Q.O J"
0 0
C7» CO
•r- CO ^3

Q. «* O 1
15 ^^
CO »-3
 CO
tt» fc« 3 •
,_ i.
O • E -C
Or- ' »-
U O E OJ "

i_« — re «/>

4-* 1 r— r—
re co re
* c it ••- 4->
CO O  >> 0
O O 4-> 1- 4J
•i— S_ •[— 3 -r-CO
to «»- co co O E
CO O CO O ~^
•i- CO Q. (1) ' — 
E c E J-  cn
•r- O
.. ^ to r— aro
co u co i — re u E
Q) re T- a> 4-> re
4J 4J E 3 O Q" 00
O CO UJ U_ r- CO re
c
4->
O r— CM CO ^- f

4-89

-------
4•5.1.2.4  Summary of emission reduction potential  of dry control  techniques  -



Table 4.11 1s a brief summary of the reductions in  NO  emissions demonstrated
                                                     A


in the preceding s^.tions.   Depending on the combustor design concept and the



type and number of control  techniques applied, the  percent reduction in NO
                                                                          A


emissions ranged from 12 to greater than 98.  Again, one must remember that



substantial effort may still be required to incorporate the techniques



demonstrated in rig tests into full turbine engines.



      Figure 4.52 shows NO  emissions for tests of  gas turbines and combustor
                          A


rigs using varying degrees and types of dry controls.



4.5.1.3  The Cumulative Effect of Dry and Met Contro] Technjques  -  The preceding



sections have quantified reductions in NO  emissions which have been demonstrated
                                         A


(either on engine tests or rig tests) by the use of wet and dry control techniques.



It has been reported, and available data support the conclusion, that dry and



wet NO  controls appear to function independently of each other and their


                       148
effects are cumulative.     This has also been demonstrated by several manu-



facturers who have incorporated some degree of dry controls into their engines



but have still needed to use water injection to meet the more stringent air



pollution codes.  Figure 4.53 presents NO  emission data for turbines using



wet plus  dry  control  techniques.


4.5.1.4  Catalytjc Exhaust Gas C1eanup  -  Environics has performed research and



development work on the catalytic reduction of NO  by NH- in a simulated gas
                                                 A      O


turbine exhaust.  Their major conclusion was that NO  reductions of 80 to 90
                                                    A

                          149 150
pe cent could be obtained. "»IJU  They report that 2000 hours of operating


                                                                                 151
time was recorded for one catalyst, with only minor deterioration in performance.



      To our knowledge, this technique has not been demonstrated on a full turbine



installation.  The laboratory results, however, look very promising,.  A unique
                                    4-90

-------
                  Table 4-11.  Percent Reduction of NO  Emissions
                               Using Dry Control Techniques
    COMPANY
                               TECHNIQUE
 PERCENT REDUCTION
       FROM
PRODUCTION COMBUSTOR
Westinghouse
Solar

General Electric -
 NASA Contract
General Electric
 (Schenectady)
                 147
Pratt & Whitney Aircraft
  NASA Contract
Westinghouse and
  Engelhard
                       Lean primary zone and reduced
                       residence time in full size
                       combustor rig

                       Lean primary in 1/2 size combustor
                       rig

                       Exhaust gas recirculation, 1/2 size
                       combustor rig

                       Vortex air blast rig test

                       Full size combustor rig tests -
                       lean burn, pre-mix, staged fuel and
                       air

                       Lean burn, fuel-air mixing
                       Rig tests; pre-mix, pre-vaporization,
                       staged combustion, lean burn


                       Catalytically supported thermal
                       combustion - rig tests
         40



         15 - 20


         30 - 38


         94<»

         51 and 60(2J



         12 - 44


         34.6 - 56.8

             60.6*

         98
(1)

(2)
This is a 94 percent reduction from the baseline VAB combustor.

Different combustor concepts
                                        4-91

-------

•





	 1 	 ~
( 	 1 	













+J 1 1531
fl cVYV- 	
01— u>
•4-* 
ep eg
1 III 11
< °» to
*^ cn V
<
5 S IJ!
2s
M
"™* *^? ?3 CD
M — _j
<
1A tt
> S £ (-
<
z
0 «- IN
a
"• «° o
(O
o
^ n *
a <* ro
< CM O
CJ  « !^ <
_i
_i
„ cvj <= fr
"> «M CO «
5
61 *** cJ UJ
U.
»- ^ m a
—. 2
_ «- 13
o
~ o E
CD
a
U, (O CM
0
CD
113

                                                                                 C
                                                                                 ^
                                                                                 (.
                                                                                 a
CN        CM        CM        »-


    ZQ lUMJsd gi i\/ Aiudd 'SNOISSIW3 XON
 .  H-
— i —i
L3  UJ
   =>
             4-92

-------
         180
         160
         140
         120
       2 100
       a.
       o.
       vf

       |  80
          61
          40
          20
                    JA-16
                    I 1.0
                    O
FACILITY CODE

   TABLE NO.

     G.T.SIZE

   FUEL TYPE
 V3

 31

17.5
22

37

33
      LIQUID (DIST.)
    V4

    32

   17.5

NATURAL GAS
     Figure 4.53. Summary of NOX emissions data from gas turbines using both dry
     and wet controls.
                                                   LEGEND
                                                   L J Combustor Rig Test
                                                   	Amount of reduction
                                                   A-l, etc., Dry Control
                                                   Technique  0.56, etc.
                                                   Water/Fuel Ratio
                                                   x unknown  water/fuel  ratio

                                                   NOTES
                                                   The lack of brackets  on  a
                                                   6.T. size  notation indicate;
                                                   a field or engine test.
                                                 4-93

-------
 advantage of this exhaust gas catalytic cleanup  device  is  that it could  be



 used for a gas turbine burning fuels with  high bound  nitrogen  contents to



 remove the NO  formed from fuel  bound  nitrogen and  thermally formed NO .



    The cost c  using this concept on a turbine has  been analyzed very



 briefly in Chapter 7 and indications are that it is not cost competitive with



 water injection or dry techniques under normal conditions.  It: is conceivable,



 -however, that further refinements of the system, production in volume or



 •application to turbines in extreme circumstances (e.g., no water) could



 {jtange the economics for this concept.



 4.5.1.5  Use of Alternate Fuels  - The  effects on emissions of  thermal NO



-burning various fuels has been discussed in  section 3.2.3.4.3.  As shown



 fn that section, a potential  control technique would  be to burn fuels which



 ^result in lower NO  emissions such as  methanol or coal  gas.  Figure 4-54 shows
                   A


 340  emissions from turbines as a function  of fuel burned.  At  present, methanol



 -«nd low Btu coal gas fuels are not commonly  available in quantity and natural



 gas supplies appear to be decreasing.



 4.5.2  Control of NO  Formed  from Fuel  Bound Nitrogen (Organic NOJ
                     ^                                          •MMMP^




    The formation of organic NO  has been discussed  in section  3.2.3.4.2.
                               A


 &n obvious control technique  for limiting  the emissions of organic NO is to
                                                                      A


 burn fuels with low bound nitrogen contents  such as natural gas or distillates.



    If high nitrogen content fuels are  used,  several potential  techniques



 exist for control of organic  NO  .  These are:                       .:
                                A


    -a.  Use of catalytic exhaust  gas cleanup  as discussed in section 4.5.1.3.



    b.  Two stage combustion with fuel  rich burning  followed by lean burning.



    The use of catalytic exhaust  gas cleanup  will  not  be discussed again  since



 ft has already been considered in section  4.5.1.3 where it was concluded that



 this concept is effective for removal  of thermal  and  organic NO ,  from the exhaust
                                                                ^'


 stream.




                                   4-94

-------
a
u
         GO
         UJ
   O
   h-
     S  "
     lU  O
     5  3
     «  <
                  o«2
                  co I-
                  h- <
              CO  **

                  -"
                     N OC
                     M O
a
O
CJ
             W  6Q

             2  5  S
             _j  >•  3
             3:  >  cc
             o  <  «
             S  UJ
-  -  s
    CO

    CD
                UJ
     O  —'     UJ  *1  X
     —  <  jf co  <  i
     JT  oc  S; O  x  £
     <  =j     oc  1-  <
     —
      <  CD
                    g  Z
                                                                                                  CM
                                                                                            U     r—
3
                                                                                                    S
                                                                                                  0  -  a
                                                                                                      CM  <*•
                                                                                                  CO     .-

                                                                                                      r~  in
                                                                                                      CM  «—
                                                                                            C9
                                                                                                «M  •-:
                                                                                                CM  «-
                                                                                                  C9
                                                                                                      cn  in
                                                                                                         CM
                                                                                                         U
                                                                                                      f  CC
                                                                                                  CO
                                                                                                      CO   .
                                                                                                      «- CM
                                                                                                      2 S
                         in
                         CM
                                                                   S
                                                    'SNOISSIW3
                                              4-95

-------
      Two stage combustion has been reported as an effective method of



oreventlnq the formation of organic NQ^152.153*154  The fuels are essentially



burned first 1n a fuel rich atmosphere and then 1n a fuel  lean atmosphere with



no conversion r ' organic nitrogen to NO .   As reported in reference 154, this



control technique has been verified for a production gas turbine by Turbodyne



Lorporation.  Based on this data, Turbodyne has calculated that emissions from



their machine when burning pure ammonia would be only 30 ppm.



4 5.3  Summary of Emission Control Systems



4.3.3.1  Thermal NO  -  As discussed 1n the preceding sections of this chapter,
                   ^


water or steam injection can be used to reduce emissions of themal NO  by
                                                                     A


percentages varying from 0 to over 80 depending only on the quantities of water



or steam injected.  Dry techniques as applied to combustor rigs; have demonstrated



the capability to reduce NO  emissions from 0 to over 90 percent, depending on



the techniques applied.  Reductions in NO  exceeding 40 percent have been
                                         A


demonstrated by application of some dry control techniques to full turbine



engines and the application of further dry controls will provide additional



reductions.



      The effects of wet and dry techniques on the reduction of NO  emissions



have been shown to be cumulative.  That 1s, NO  emissions from a given engine



may be reduced by 40 percent through the application of dry techniques and then



further reduced by applying wet techniques.



      Therefore, the level of an emission standard for thermal NO  will not be
                                                                 A


determined by the limit of technology since almost any level cam be obtained



by simply injecting additional quantities of water or using  combinations  of wet


 and dry control s_.	



 4.5.3.2  Organic N0y -  Laboratory tests and tests on boilers  have shown  that



 catalysts used in exhaust gas  streams  are  over 70 percent effective  1n  removing




                                    4-96

-------
NO , regardless of source.  Also, as discussed earlier 1n this chapter, two
  X


stage combustion (rich combustion followed by lean combustion) can be used



to eliminate formation of organic NO .  This has been demonstrated 1n a
                                    n


production turbine engine marketed by Turbodyne Corporation.  A third technique



is to burn low nitrogen fuels.



      In any case, gas turbines do not commonly burn the high nitrogen fuels.



Also, as discussed in section 4.3, there does not appear to be any present



trend towards the combustion of those fuels in gas turbines.  Since the



contribution of organic NO  from a turbine burning a high nitrogen fuel can
                          /\


exceed the thermal NO , a gas turbine regulation to control NO  emissions should
                     y\                                        ^


regulate both forms of NO .
                         A


4.5.3.3  HC and CO  -  As discussed 1n Chapters 3 and 4, HC and CO emissions



from gas turbines are not significant for full load operations.  However,



emissions of both pollutants increase with decreasing load.  Wet control



techniques do not reduce CO or HC emissions.  Several dry control techniques,



however, have been demonstrated for reducing the CO and HC emissions over the



turbine operating range.



      Most gas turbines are operated only at or near full load conditions.



Some utilities, however, operate gas turbines tn a low power "spinning reserve"



mode so that they can be rapidly brought up to full power 1n case of frequency,



voltage or power fluctuations.  For example, a 67 MW turbine operating 1n



spinning reserve emits about 960 pounds per hour of CO.  Since turbines are



commonly used in clusters, it is obvious that a potential exists for significant



impact on local ambient air quality.  This impact is considered in Chapter 6.



4.5.3.4  SOp -  As discussed in section 4.3, the emissions of SO^ from gas



turbines are strictly a function of the sulfur content of the fuel burned



and flue gas scrubbing for S02 removal is impractical (compared to the cost



of desulfurizing the fuel) because of the large volumes of gas which must be




                                     4-97

-------
treated.  The relative costs are briefly discussed 1n Chapter 7.  Therefore,
a standard for S02 emissions would essentially limit the sulfur content
of the fuel which car be burned and would depend on factors other than control
techniques applicable to the turbine.
                                     4-98

-------
                      REFERENCES FOR CHAPTER 4

 1.   Johnson, R.  H., and C. Wllkes.  "Environmental  Performance of Industrial
     Gas Turbines".   General Electric.   ASME Publication 74-GT-23.  pp 4-6.
     April 4, 1974.
 2.   Johnson, R.  H., and F. C. Wilhelm.  "Control  of Gas Turbine Emissions in  the Worli
     Environment".   General Electric,   p 14.  Copyright by General Electric  Co.   1974
 3.   Lee, S. Y.,  et al.   "Evaluation of Additives  for Prevention of High Temperature
     Corrosion of Superalloys in Gas Turbines". Westinghouse.   pp 1-7.  ASME
     Publication  73-GT-l.
 4.   Ambrose, M.  J., and E. S. Obidinski.  "Recent Field Tests  for Control of
     Exhaust  Emissions from a 35-MW Gas Turbine".  Westinghouee.  ASNE Publication
     72-JPG-GT2.   p 5.  September 1972.
 5.   Op.  Cit., Reference 1.
 6.   Carl, D. E., et al.  "Exhaust Emissions from  a 25-MW Gas Turbine Firing,  Heavy
     and Light Distillate Fuel Oils and Natural Gas".  Hestinghouse.  ASME Publication
     75-GT-68.  pp8-ll.  December 2,  1974.
 7.   Appendix to  GASL-TR-787.  "Gas Turbine Emission Measurement Program" for
     Empire State Energy Research Corporation, New York, New York.  "Turbine
     Exhaust Emissions Measured at Facilities of the New York Power Pool" prepared
     by Scott Research Laboratories, Inc.   pp 1-26.  March 1974.
 8.   Letter and attachments from Carole Kleinman,  Teller Environmental Systems,  Inc.,
     to Mr. Don R.  Goodwin, EPA, ESED,  March 16, 1973.
 9.   Letter and attachments from Stuffert, C. F.,  Jr., Teller Environmental  Systems,
     Inc. to B. J.  Steigerwald, EPA.  January 27-,  1975.
10.   Plonsker, et al.   "Reduction of Gas Turbine Smoke and Particulate Emissions by
     a Manganese  Fuel  Additive".  Presented at the Central States Section of the
     Combustion Institute,   pp 1-12.  March 26-27, 1974.
                                    4-99

-------
11.   "Discussion of the Paper by  Plonsker,  et al  -  Reduction of Gas  Turbine
     Smoke and Particulate  Emissions  by  a Manganese Fuel Additive".   Presented
     at the Central  States  Section  of the Combustion Institute,   pp  1-8.
     March 26-27,  1974.
12.   Decorso,  S. M., et al.   "Smokeless  Combustion  in Oil Burning Gas Turbines".
     ASME Publication 67-PWR-5.   pp 1-8.  July  7, 1967.
13.   Bahr, D.  W., and C.  C.  Gleason.   "Technology for the Reduction  of Aircraft
     Turbine Engine Pollutant Emissions".   ICAS Paper Number 74-31.   pp 3-4.
     August 1974.
14.   Op.  CH., Reference 1,  p.  7
15.   Op.  Cit., Reference 13, p. 13
16.   Op.  Cit., Reference 14, p. 7
                            ^
17.   Op.  Cit., Reference 2,%p.  15-16
18.   "General  Motors Response to  Preliminary (Draft) Proposed  Standards for
     Control of Air Pollution From  Stationary Gas Turbines".   Submitted to EPA,
     ESED, by GM Environmental  Activities Staff,  pp 24, 34.   March  21, 1973.
19.   "Advanced Combustion Systems for Stationary Gas Turbines".   Proposal
     prepared by General Electric in response to EPA, RFP Number  DU-75-A182;
     May 9, 1975.  Transmitted to Dr. B.  Steigerwald, EPA,  OAQPS, via letter
     from D. R. Plumley, GE.  May 20, 1975.  p  26.
20.   Op.  Cit.,  Reference 12, p.  6-7
1.1.   Op.  Cit.,  Reference 12, p.  6
22.   Letter and attachments from  Zeltmann,  E. W.  (General Electric)  to D.  Walters,
     EPA.  June 15, 1973.
23.   "Mineral  Industry Survey".   U. S. Department of Interior, Bureau of Mines,
     Petroleum Products Survey Number 76, Burner Fuel Oils.   1972.
                                  4-100

-------
24.   Op. Cit., Reference 17, p.  10
25.   Letter and attachments from Decorso, S.  M.  (Westinghouse) to D.  R.
     Goodwin, EPA, ESED.  pp 26, 28.   January 8, 1976.
26.   Letter from Mr.  Boney, J.  W., Jacksonville  Electric Authority, to D.  R.
     Goodwin, EPA.  p 2.  November 13, 1975.
27.   Letter and attachments from Assard, D.  G.,  Turbo Power and Marine,  to
     D. R. Goodwin, EPA.  pp 6-7.  November 26,  1975.
28.   Letter and attachments from Hoppe, P.,  ASME Gas Turbine Division, to  D.  R.
     Goodwin, EPA.  October 9,  1975.
29.   Responses from 17 electric utilities submitted by Baruch, S. B., Edison
     Electric Institute, to K.  R. Durkee, EPA.  October 1975 - January 1976.
30.   Op. Cit., Reference 24.
31.   Durkee, K. R.  (EPA)  Turbo Power and Marine Meeting Report,  p  4.
     May 4, 1976.
32.   Op. Cit., Reference 13, p.  4-8.
33.   Letter and attachments from Medigovich,  D.  G., Garrett Aireasearch, to
     D. R. Goodwin, EPA.  Appendix I.  p 13.   February 13, 1976.
34.    Op.  Cit., Reference 13, p. 16.
35.    Op.  Cit., Reference 13, p. 18
36.    Op.  Cit., Reference 13, p. 17.
37.   Bahr, D. W., and G. C. Gleason.   General Electric.  "Experimental Clean
     Combustor Program".  Phase I Final Report.   Prepared for NASA, NASACR134737.
     p 148.  June 1975.
38.   Roberts, R., et al.  Pratt and Whitney.   "Experimental Clean Combustor
     Program".  Phase I, Final  Report.  Prepared for NASA, NASACR-134736,  PWA
     5153.  pp 86-97.  October 1975.
39.    Op.  Cit., Reference 37, p. 26
                                    4-101

-------
40.   Op. C1t., Reference 37, p.26
41.   Hung, W. S. Y.  "The Reduction of NO  Emissions from Industrial  Gas
                                         ^
     Turbines",  llth International Congress on Combustion Engines,,   Barcelona.
     p 5.  1975.
42.   Niedzwiecki, R. W., and R.  E.  Jones.  "The Experimental  Clean Combustor
     Program - Description and Status."  NASA Technical  Memorandum NASA TMX-71547
     (NTIS:N74-21399).   pp 5-6.   May 2, 1974.
43.   Op. Cit., Reference 37, p.  8-9.
44.   Johnson, R. H., and C. Wllkes.  "Environmental  Performance of Industrial
     Gas Turbines".  ASME Publication 74-GT-23.  pp 6-7.  April 4, 1974.
45.   Cornelius, W., and W. R. Wade.  "The Formation and Control of Nitric Oxide
     in a Regenerative Gas Turbine".  SAE Paper 700708.   pp 10-14,  September
     1970.
46.   Letter  and attachments from Gaylord, R. H., Turbodyne Corporation, to
     D. R. Goodwin, EPA.  Section 3.  December 19, 1975.
47.    Op. Cit., Reference 19, p. 35.
48.   Mularz, E. J., et al.  "Pollution Emissions from Single Swirl Can
     combustor Modules at Parametric Test Conditions".  NASA TMX-3167.  p 2.
     January 1975.
49.   Op. Cit., Reference 4, p. 1-9.
50.   Op. Cit., Reference 18, p.  26-31.
51.  Dp. Cit., Reference 33, p.  3-41 through 3-46.
52.   Kollrock,  R.,  and L. D. Aceto.  Pratt and Whitney.   "The  Effects of
     Liquid  Water Addition  in Gas Turbine Combustors".  Journal of the Air
     Pollution  Control Association.  Volume  23, Number  2.  pp  116-121.
     February  1973.
53.   Klapatch,  R. D., and T. R. Koblish.  "Nitrogen Oxide Control with Water
     Injection  in Gas Turbines".  ASME Publication 71-WA/GT-9.  pp 1-8.
     August  1971.
                                    4-102

-------
54.   Letter from Dibelius, N.  R.,  Vice Chairman,  ASME Combustion and Fuels
     Committee to Steigerwald, B.  J.,  EPA.   Enclosure 3.   May 1, 1973.
55.   Op.  Cit., Reference 46  Section F
56.   Op.  Cit., Reference 27, p.  6
57.   Letter and enclosures from Zeltmann, E., General Electric Corporation,
     to K. R. Durkee, EPA.  p 31.   October 31, 1975.
58.   Ibid, Reference 57.  p 31.
59.   Letter from Dietz, J. F., San Diego Gas and  Electric Company,  to D.  Bell,
     EPA.   January 31, 1975.
60.   Op.  Cit., Reference 25, p.  15-16.
61.   Op.  Cit., Reference 57, p.  68-70.
62.   Op.  Cit., Reference 57, p.  68-70.
63.   Op.  Cit., Reference 46  Section D
64.   Op.  Cit., Reference 27  Appendix  22
65.   Featherston, C. H.  "Retrofit Steam Injection for Increased Output".
     Gas Turbine International,   pp 32-35.   June  1975.
66.   Ibid, Reference 65.  pp 32-35.
67.   Op.  Cit,, Reference 46  Section D
68.   Op.  Cit., Reference 27  Appendix  22
69,   Op.  Cit., Reference 27, Attachment 26
70.   Op.  Cit., Reference 25, p.  24
71.   Letter from Zitlan, W. A.,  President, San Diego Gas  and Electric Company,
     to D. R. Goodwin, EPA, ESED.   p 5.  October  19, 1972.
72.   Vogel, R. G., Southern California Gas Company.  "Analysis of EPA Suggested
     New Source Performance Standards for Stationary Gas  Turbines",  pp 6-7.
     March 1975.
                                 4-103*

-------
73.  "Economic Impact of New Source Performance Standards  on Stationary
     Gas Turbines".  Prepared by Energy and Environmental  Analysis,  Inc.,
     for EPA.  Contract Number 68-02-2082.   p III-6.
74.  Ahner, D. J.  (General Electric).   "Environmental  Performance".   Copyright
     1971 by General Electric Company,   p 5.
75.  Ibid, Reference 74.  p 7.
76.  CD. Cit., Reference 27, p. 3-4
77.  Op. Cit., Reference 71, p. 6-7 Figure A-l
78.  Letter and attachments from Paraskeva, G.  C.,  Basin Electric Power Corporation,
     to R. Train, Administrator, EPA.  October 3,  1975.
79.  Letter and attachments from Snyder, R. B., Portland General Electric
     Company, to K. R. Durkee, EPA.   Attachment 1.  December 26, 1975.
80.  "Preliminary Draft Report - Economic Impact of New Source Performance
     Standards on Stationary Gas Turbines".  Prepared for EPA by Energy and Environ-
     mental Analysis, Inc.  Contract Number 68-02-2082.  pp III-7 through III-9.
     December 2, 1975.
81.  Op. Cit., Reference 4, p. 1717.
82.  Lipfert, F. W., et al.   "The New York Power Pool Gas Turbine Emissions
     Test Program".  Paper presented at the APCA Conference, Kiamesha Lake, New
     York,  pp 8-10.  October 13-15, 1974.
83.  Op. Cit., Reference 18,  pp.27-28, 40-50.
84.  "Comments of the General Electric Gas Turbine Products Division on the
     Impact of Fuel-Bound Nitrogen on the Formation of Oxides of Nitrogen from
     Gas Turbines".  Based on work by Johnson and Wilkes, Edited by Hamilton and
     Zeltman.  pp 1-25.  January 28, 1974.
85.  Ibid, Reference 84.  p 21.
86.  Op. Cit., Reference 84,  p. 20
                                   4-104

-------
 87.  Op.  Cit.,  Reference 46, Section G.3.5
 88.  Op.  Cit.,  Reference 27, p.  31
 89.  Letter, Furney, J. P., Pacific Gas and Electric, to Baruch, S.  B.,
      Edison Electric Institute,   p 6.  November 4, 1975.
 90.  Ohtake, T.  "Studies on Ice Fog".  Public Health Service Report AP-00449.
      pp 1-17.   June 1970.
 91.  Niedgwiecki, R. W., and R.  E. Jones.  "The Experimental  Clean Combustor
      Program - Description and Status".  NASA TMX-71547.  Table IV.   May 1974.
 92.  Op.  Cit.,  Reference 18, p.  39
 93.  Op.  Cit.,  Reference 18, p.  53
 94.  Op.  Cit.,  Reference 13, p.  54
 95.  Op.  Cit.,  Reference 13, p.  21
 96.  Op.  Cit.,  Reference 13, p.  21
 97.  Singh, P.  P.,  et al.  "Formation and Control  of Oxides of Nitrogen Emissions
      from Gas  Turbine Combustion Systems".  ASME Paper Number 72-6T-22.   p  4.
      March 1972.
 98.  Ibid, Reference 97, "p. 4-5
 99.  Op.  Cit.,  Reference 97, p.  6
TOO.  Op.  Cit.,  Reference 97, p.  6
101.  Op.  Cit.,  Reference 97, p.  5-7
102.  Roberts,  P.  B., et al.  (Solar)  "Advanced Low NO  Combustors for  Supersonic
                                                       A
      High Altitude  Aircraft Gas  Turbines".  ASME Publication  76-GT-12.   pp  1-19.
103.  Ibid, Reference 102, p. 3
104.  Op.  Cit.,  Reference 102, p.  4
105.  Op.  Cit.,  Reference 102, p.  5
106.  Op.  Cit.,  Reference 102, p.  6
107.  Op.  Cit.,  Reference "102, p.  6
                                     4-105

-------
108.   Op.  Cit.,  Reference  102,  p.  6
109.   Op.  Cit.,  R^ierence  102,  pp. 7-10
110.   Op.  Cit.,  Reference  102,  p.l
111.   Op.  Cit.,  Reference  102,  p.  18
112.   Op.  Cit.,  Reference  102,  p.  3
113.   Op.  Cit.,  Reference  102,  pp. 12-14
114.   Op.  Cit.,  Reference  102,  pp. 15-18
115.   Op.  Cit.,  Reference  102,  pp. 17-18
116.   Bahr, D.  W., and C.  C.  Gleason.   General  Electric  Corporation.
      "Experimental  Clean  Combustor  Program,  Phase  I  Final  Report".   NASACR
      134737.   p 18.   June 1975.
117.   Ibid, Reference 116.  p 19.
118.   Op.  Cit.,  Reference  116,  p.  83
119.   Op.  Cit.,  Reference  116,  p.  62
120.   Op.  Cit.,  Reference  116,  p.  86
121.   Op.  Cit.,  Reference  116,  p.  93
122.   Op.  Cit.,  Reference  116,  p. 63
123.   Op.  Cit.,  Reference  116,  p.  66
124.   Op.  Cit.,  Reference  116,  p.  145
125.   Op.  Cit.,  Reference  116,  p.  152
126.   Roberts,  R., and A.  Peduzzi  and G.  E.  Vittl.   Pratt and Whitney Aircraft
      Division.   "Experimental  Clean Combustor  Program - Phase I  Final  Report".
      NASACR-134736.   PWA 5153.  p 40.   October 1975.
127.   Ibid, Reference 126.  p 51.
128.   Op.  Cit.,  Reference  126, p. 60
                                     4-106

-------
1?9.  Op. Cit., Reference 126, p.  99.
,30.  Carruba,  R.  V., et al.   Engelhard Industries.   "Catalytlcally Supported
      Thermal Combustion for Emission  Control".   Paper presented  at the  Electric
      Power Research Institute NO   Control  Technology Seminar,  pp.  3-8.
                                /\
      February  6,  1976.
131.  Decorso,  S.  M., et al.   "Catalysts for Gas  Turbine Combustors -  Experimental
      Test Results".  ASME Paper Number 76-GT-4.   pp. 1-9.   November 14,  1975.
132.  Ibid, Reference 131.  p. 2.
133.  Op. Cit., Reference 131, pp. 1-9.
134.  Op. Cit., Reference 130, p.  3.
135.  Pffefferle,  W. C., et al.  Engelhard  Industries.   "Catathermal Combustion:
      A new Process for Low Emissions  Fuel  Conversion".   ASME  Publication 75-WA/
      FU-1.  pp.  1-13.
136.  Blazowski,  W. S., and G. E.  Bresowar.  Air  Force Aero Propulsion Laboratory.
      "Preliminary Study of the Catalytic Combustor Concept as  Applied to Aircraft
       Gas Turbines".   AFAPL-TR-74-32.   pp. 10-15.   May  1974.
137.  Op. Cit., Reference 131.
138.  Op. Cit., Reference 131, p.  3.
139.  Op. Cit., Reference 131.  p. 4.
140.  Wampler,  F.  B., et al.   "Catalytic Combustion of C,HQ on  P.  Coated  Monolith".
                                                        O O     L
      The Combustion Institute Paper  Number 74-36.   pp.  1-25.   October 1974.
141.  Anderson, D. N.,  et al.   "Performance of a  Catalytic Reactor at  Simulated
      Gas Turbine Combustor Operating  Conditions."   NASA TMX 71747.  pp.  1-19.
      June 1975.
142.  Op. Cit., Reference 134.  pp. 1-53.
143.  Op. Cit., Reference 130.  Table  I.
144.  Op. Cit., Reference 130.  p. 14.
145.  Op. Cit., Reference 130.  p. 15.

                                    4-107

-------
146.   Durkee, K.  R.   EPA.   "Meeting  with  Personnel  from  Engelhard  Industries
      Division",   p.  1-2.   May 7,  1976.
147.   General Elect"  c, Letter and Attachments  from E. W.  Zeltmann,  to  K.  R.
      Durkee, EPA, ESED.   pp.  26-29.  October 31,  1975.
148.   Singh, P.  P.,  et al.   "Formation and Control  of Oxides  of Nitrogen
      Emissions  from Gas  Turbine Combustion Systems." ASME Publication 72-GT-22.
      ij.  4.  December 1972.
149.   Kiine, 0.  M.,  et al.   Environics, Incorporated. "Catalytic  Removal  of
      Nitrogen Oxides:  Application  to Gas Turbine Exhaust".   Summary Report
      73-5001-8.   (Undated but received in 1973).
150.   Kline, J.  M.,  Environics, Incorporated.  "Preliminary  Design Study:
      Prototype NO  Reducing Catalytic Converter for Gas Turbines."   Report
                  /\
      Number 73-5001-8.  (Undated but received in  1973).
151.   Op. Cit., Reference 149.  p. 1.1.
152.   Turner, D.  W., et al.  Esso Research and Engineering.   "Influence of
      Combustion Modification and Fuel Nitrogen Content  on Nitrogen Oxides
      Emissions from Fuel Oil  Combustion".  Combustion,   pp.  21-29.   August 1972.
153.   Op. Cit., Reference 19, p. 19.
154.   Op. Cit., Reference 46.  Section 6.3,3.  pp. 1-10.
                                      4-108

-------
                   5.   MODIFICATION AND RECONSTRUCTION

     In accordance with Section 111  of the Clean A1r Act, as  amended  in
1970 and 1974, standards of performance shall  be established  for new
sources within a stationary source category which ". . .  may  contribute
significantly to air pollution ..."  Standards apply to operations  or
apparatus (facilities) within a stationary source, selected as "affected
facilities", that is,  facilities for which applicable standards of
performance have been  promulgated and the construction or modification
of which commenced after the proposal of said  standards.
     On December 16, 1975, the Agency promulgated amendments  to the
general provisions of 40 CFR Part 60, including additions and revisions
to clarify modification and the addition of a  reconstruction  provision.
Under the provisions of 40 CFR 60.14 and 60.15, an "existing  facility"
may become subject to standards of performance if deemed  modified or
reconstructed.  An "existing facility" defined 1n 40 CFR  60.2(aa) is  an
apparatus of the type for which a standard of  performance is  promulgated
and the construction or modification of which  was commenced before the
date of proposal of that standard.  The following discussion  examines
the applicability of these provisions to stationary gas turbine facilities
and details conditions under which existing facilities could  become subject
to standards of performance.  It is important  to stress that  since standards
of performance apply to affected facilities, which, combined  with existing
and other facilities comprise a stationary source, the addition of an affected
facility to a stationary source through any mechanism, new construction,
                                    5-1

-------
modification, or reconstruction, does not make the entire stationary source
subject to standards of performance, only the added affected facility.
5.1  40 CFR PART 63 PROVISIONS FOR MODIFICATION AND RECONSTRUCTION
5.1.1  Modification
     It is important that these provisions be fully understood prior to
investigating their applicability.
     Section 60.14 defines modification as follows:
          "Except as provided under paragraphs (d), (e)  and (f) of
     this section, any physical or operational changes to an existing
     facility which result in an increase in emission rate to the
     atmosphere of any pollutant to which a standard applies shall
     be a modification.  Upon modification, an existing facility shall
     become an affected facility for each pollutant to which a standard
     applies and for which there is an increase in the emission rate".
     The exception in paragraph (d) allows for an existing facility to
undergo a physical or operational change which results in an increase in
the emission rate of any pollutant to which a standard applies, but not
be deemed a modification, provided the owner or operator can demonstrate
to the Administrator's satisfaction that the total emission rate of that
pollutant from all facilities within the stationary source to which an
appropriate EPA reference method or alternative method listed in paragraph
(b) can be applied has not increased.  The required reduction in emission
rate may be accomplished through the installation or improvement of a
control system or through physical or operational changes to facilities
including reducing the production of a facility or closing a facility.
     Paragraph (e) lists certain physical or operational changes which
will not be considered as modifications, irrespective of any change in
the emission rate.  These changes include:
                                    5-2

-------
     1 - Routine maintenance, repair, and replacement.
     2 - An increase In the production rate not requiring a capital
expenditure as defined in Section 60.2(bb).
     3 - An increase in the hours of operation.
     4 - Use of an alternative fuel or raw material  if prior to the
standard, the existing facility was designed to accomodate that alternate
fuel or raw material.
     5 - The addition or use of any system or device whose primary function
Is the reduction or air pollutants, except when an emission control  system
Is removed or replaced by a system considered to be less efficient.
     Paragraph (b) clarifies what constitutes an increase in emissions in
kilograms per hour and the methods for determining the increase, including
the use of emission factors, material balances, continuous monitoring
systems, and manual emission tests.  Paragraph (c) affirms that the  addition
of an affected facility to a stationary source does not make any other
facility within that source subject to standards of performance.  Paragraph
(f) simply provides for superceding any conflicting provisions.
5.1.2  Reconstruction
     Section 60.15 regarding reconstruction states:
          "If an owner or operator of an existing facility proposes
     to replace components, and the fixed capital cost of the new
     components exceeds 50 percent of the fixed capital cost that
     would be required to construct a comparable entirely new
     facility, he shall notify the Administrator of the proposed
     replacements.  The notice must be postmarked 60 days (or as
     soon as practicable) before construction of the replacements
     is commenced. . ."
     The purpose of this provision is to ensure that an owner or operator
does not perpetuate an existing facility by replacing all but vestigial
components, support structures, frames, housings, etc., rather than  totally

                                  5-3

-------
replacing it in order to avoid subjugation  to  applicable standards of
performance.
5.2  APPLICABILITY TO STATIONARY  GAS  TURBINE INSTALLATIONS
5.2.1  General1' 2* 3» 4
        In the gas turbine Industry,  the  terms "overhaul" and  "ultimate  life"
are not definable In terms used equally by  all manufacturers.   As an example,
one manufacturer's "overhaul"  is  another  manufacturer's  "major disassembly
Inspection".  To compound this nebulous situation,  the intervals called
"overhaul life" and "ultimate  life" vary  significantly among manufacturers
and users.  The variations in  "overhaul life"  and "ultimate life" are
Influenced by differences in gas  turbine  applications, operating loads,
operating hours per start, fuel quality and maintenance  practices.  Most
turbines are designed for 20,000  to 40,000  operating hours  between overhauls
and for an ultimate life of about 30 years. The 30 year figure for ultimate
life is the design life of non-wearing parts such as rotor  discs, casings,
housing, bearing supports, and rotor shafts.   Typically  an  overhaul returns
a unit to a "new" condition through the replacement of wearing or working parts
and, therefore, the "ultimate  life" of a  gas turbine may be indefinite.
5.2.2  Modification
        Physical and operational  changes  to a  gas turbine which might  be
considered a modification are:
        a.  Replacement of components with  a different design  than the original
to permit firing a turbine with fuels for which it was not  originally  designed,
si  ;h as crude or residual oils.
                               5-4

-------
     b.  Replacement of components with a different design than the original
to increase the power output of the turbine.
     c.  Sustained operation of a turbine at higher outputs than design
(this can be performed at a sacrifice in turbine life).
     It is highly unlikely that an owner or operator would opt to perform
items a, b, or c above because:
     a.  As discussed in Chapter 4, there does not appear to be any trend
to use crude or residual oils in gas turbines.
     b.  Power output of a turbine is commonly increased by water or steam
injection rather than replacement of components
     c.  Gas turbine life decreases rapidly when operation occurs above
design levels for sustained periods.
     As indicated in Chapter 3, stationary gas turbines have many uses
and may be located at oil refineries, power plants, pumping stations, etc.
When a gas turbine is located in a plant which has other sources of the
same pollutants generated by turbines (e.g., refineries, power plant
complexes, etc.), then increased pollutants due to a modification of the
turbine can be offset by the compensatory emission reduction allowed under
paragraph (d) of Section 60.14.  This compensatory emission reduction must
be accomplished by upgrading an existing control system, by adding a new
control system, or by making physical or operational changes on facilities
within the source which are amenable to conventional emission tests methods
so that the owner or operator can demonstrate by emission tests that no
net increase in emissions has occurred.  The exception to the emission
testing requirement would occur if the compensatory emission reduction
resulted from an operational change such as reducing the production capacity
or shutting down a piece of process equipment to which an acceptable emission
factor applied.
                                    5-5

-------
     The following physical or operational changes will  not be considered
as modifications to existing gas turbines:
     a.  Changes deteimined to be routine maintenance, repair, or replacenent
in kind.  This will include repair or replacement of stator blades, turbine
nozzles, turbine buckets, fuel nozzles, combustion chambers, seals, and shaft
packings.
     b.  Changes in type or grade of fuel used, if the original qas turbine
installation, fuel nozzles, etc., were designed for its  use.
     c.  An increase in hours of operation.
     d.  Variations in operating loads within the engine design specification.
As noted in Chapter 3, emissions from gas turbines vary  with operating loads.
     e.  The relocation or change in ownership of an existing facility.
     The impact of the modification provision on existing gas turbines should
be very slight.
5.2.3  Reconstruction
     A reconstructed turbine, as discussed in 5.1.2, is  essentially a turbine
which has undergone a major rebuilding when  it would otherwise have been
scrapped or recycled.  It is difficult to apply the definition of reconstruction
to a gas turbine because substantial portions of a turbine may be replaced as
a matter of routine maintenance during the normal overhauls as described in
5.2.1.  Since it is current practice to replace substantial portions of turbines,
it would be difficult to discriminate between a major overhaul that was performed
to avoid the purchase of a new turbine and one that was  performed in accordance
with a routine maintenance program.  Such routine maintenance should be exempted
from the regulatory consequences of becoming a reconstructed turbine, subject
to the "50 percent rule" discussed in 5.1.2.

-------
                         REFERENCES FOR CHAPTER  5

1.  Letter and attachments, Sievert, M. 0.,  Solar,  to  Don  R.  Goodwin,
    EPA, ESED.  p.  4.   November 10, 1975.
2.  Letter and attachments, Gaylord, R. H.,  Turbodyne, to  Don R. Goodwin,
    EPA, ESED,  Section C,  December 19, 1975.
3.  Letter and attachments, Zeltmann, E.,  General Electric,  to Kenneth R.
    Durkee, EPA, ESED.   p. 22.
4.  "Gas Turbine Maintenance".   R.  H. Knorr.  General  Electric,  pp. 1-13.
    1974.
                                    5-7

-------
                           6.  ENVIRONMENTAL IMPACT

     As discussed in section 3.2.3.5, only a few state and local  jurisdictions
have regulations specifically for stationary gas turbines, although some have
applied standards originally written for other sources to gas turbines.
Therefore, for purposes of this chapter, the environmental impact of the
emission control techniques will be discussed by comparing their Impact with
the impact assuming gas turbines are uncontrolled.
6,1  AIR POLLUTION IMPACT
6.1.1  Introduction
     This section evaluates the air quality Impact of different levels
of control for S09, CO, and NO  emissions from gas turbines.  Through the use
                 £            A
of dispersion models, concentration estimates are made for each of the three
pollutants for averaging times consistent with National Ambient A1r Quality
Standards (NAAQS).   These standards and averaging times are tabulated below
for reference.
Pollutant
N02
CO

so2


Averaging Time
Annual
8 hr.
1 hr.
Annual
24 hr.
3 hr,
Standard \
Primary
100
10 mg/m
40 mg/m
80
365

ig/m
Secondary
100
10 mg/m
3
40 mg/m


1300
                                        6-1

-------
     Dispersion analyses for nine stationary gas turbines operating singly



and in clusters and for operation at full  load and "spinning reserve" are



considered.  For these analyses, the effects of varying the sulfur content of



the fuil from 0.01 to 2.7 percent are investigated.   Uncontrolled levels of



CO emissions and alternative controlled levels of 90 ppm, 125 ppm, 250 ppm



and the estimated uncontrolled level at spinning reserve of 500 ppm are



analyzed.  The effect on ambient air quality of uncontrolled levels of NO
                                                                         A


emissions and alternative controlled levels of 75 ppm, 100 ppm, and 125 ppm



is also analyzed.  As discussed in Chapter 3 and 4:   (a)  S02 emissions are



controlled by limiting the sulfur content of the fuel used; (b)  CO emissions



are a function of combustor geometry and turbine efficiency; (c)  NO  can be
                                                                    f\


controlled to almost any level by water injection and dry control techniques.



     The dispersion analyses also considers the effects on ambient air quality



levels of varying wind speeds and variations in exhaust stack height.



     For the purposes of these analyses, all three pollutants are assumed to



behave in the atmosphere as non-reactive gases.  All NO  emissions and concen-
                                                       n


trations are assumed to be equivalent to those for N0«.  The estimated pollutant



concentrations are based on the application of state-of-the-art modeling



techniques, which implies reliability of the estimates to within a factor or two.



6.1.2  Source Characterjsties



     The major emphasis in these analyses is an assessment of the maximum



potential air quality impact of S0?, CO, and NO  emissions for both  individual
                                  C~            A


ar  cluster arrangements of gas turbines.  Source data for the rttne  individual



stationary gas turbine units are reported in Table 6-1.  The difference in emission



levels for the gas turbine unit is proportional to their levels of power output.



Thus, it can be expected that the individual air quality impact for  the larger



gas turbine units may be greatest.





                                   6-2

-------
Manufacturer &
       Table 6 -  1
      Stack Parameters
Power       Cycle Type  H
Model Output

(MW)

s
(•)
s
tteg.K)
s
s
(m/sec) (m)
"b
(m)
Garret Ai research
1.

GTC8S-90 0.158
(GPUT-90)
Simple

1.9

903

57

0.3

5

Solar
2.
3.
Saturn 0.8
Centaur 2.5
Simple
Simple
4.6
4.9
733
700
77
68
0.45
0.75
3.69
3.38
General Electric
4.

5.
6.

7.




8.


9.

MS5001P 24.2
(Spinning Reserve Mode)
MS5002B 23.5
MS7001B 60.3
(Spinning Reserve Mode)
MS7001B 161.6
(Two Machines)



MS7001B 57.4

(Spinning Reserve Mode)
MS7001C 67.4
(Spinning Reserve Mode)
Simple

Simple
Simple
•
Unflred
Combined

By-Pass
Stack
Regenera-
tive

Simple

10.8

6.4
9.7

15.9


12.8
754
(510)
622
780
(528)
444


780
19.2
(12.2)
15.7
33.7
(20.7)
27.4


0.38
4.12

3.89
4.44

5.33
(Double
Unit)
4.31
7.9

7.9
8.69

8.69



(Each Unit)
9.4


9.7

639

(432)
811
(549)
23.7

(15.1)
38.5
(24.5)
3.44


4.46

8.69


8.69

    HS = stack height
    TS = stack temperature
    Vs = stack exit speed
    DS = equivalent stack diameter
    H.  = height of highest nearby structure
                                           6-3

-------
     Turbine units number 4 through 9 have rectangular shaped stacks
so the diameter *s based on an equivalent area circle.  Stack velocity
is based on the mass flow rate and stack exit area.   Turbine units
number 4, 6, 8, and 9 additionally are considered 1n a spinning
reserve mode where the exit temperature and velocity are reduced.
     The cluster arrangements for which analyses are made are shown
in Figure 6-1.  These arrangements represent expected groupings of
the gas turbines.

6.1.3  Dispersion Models
     The dilution of effluents from stationary gas turbine units 1s
much different from that of sources with conventional tall round
stacks.  For many gas turbines the effluent stacks are not much
higher than the nearby building structure which houses the turbines.
The short effluent stacks may adversely affect both plume rise and
dispersion.  In addition, many of the effluent stacks are unconven-
tionally rectangular in shape and some are stuffed with acoustic
baffling which results in a nonuniform effluent exit stream.  Under
high wind conditions these stack characteristics can be expected to
frequently  result 1n a severe effect on dispersion of the effluent
due to entrainment of the plume into the stack wake.  The adverse
affects of  the turbine structure and stacks on plume rise and
effluent dispersion generally result in higher concentrations than
those associated with a source having similar emission rates, but
more conventional characteristics.
                              6-4

-------
GTC85 90(G PUT-90)
MS5001BANOMS5002B
MS7001B(UNFIREO
.COMBINED CYCLE)
16
O


11
O



6
O

6







MS7001
i










17 18 19 20
0 O 0 Oi
L^-SOnvW 1
100m
12 13 14 15 1
O O O CH



7 8 9 10
O O O O

2345
OO.OO
» •» "i
— r" °
5 « » o
(A



B (REGENERATIVE CYCLE)
15m— *-| |-«-
Ib O O16b
-i-i A...
n~rQ Qisb
7i 6 6l5a
8b-r-0 Ol4b
i-i 6 14a
Sb3°mO Ol3b
«• O Ol3a
4b 0 Ol2b
4i O Ol2a
3b O Ollb
3a O Dili
2b O OlOb
*• 6 OlOa
1b O O 9b
It O O 9a
_. 1 aj
x>! "
— r- ! °
>|t|
B ao
r ••
•Or
10m
'O1-
•o
SO

•o

'O
«o
•o
l> 1

> ° 5
BSyl
-



SATURN AND
U— 10m


•O
- 'O
•o
•o.
«o
'O
'O
:Q
~t _^ I *H
>< ot o
R s
ram 	















•


CENTAUR
H
o»
.o»
O M
O"
o»
O"
O"
O's
o»
O"
o»

•
*l
iiQ

«o
«o
»o

"O

"O
"O
•o



















15M T-g 16M

ft 14BP
13M -T-O 14M
13BP f o
To 12BP
11M J-0 12M
11BP o
§10BP
10M
...
§8BP
8M
—
o 6BP
5M A 6M
5BP 0
§4BP
4M
__.
1M 8 2M
IBP 9-0
•4 ~^ i f*
X > 0
BO! "
*gii
B t =
(M) MAIN STACK
(BP) BY-PASS STACK
MS7001B (SIMPLE CYCLE)
MS7001C
nQ-L
>-C30m
14 fjT
13W
I2W
ioQ
•p
•O
sQ
jo
'0.
^C ^k i LK
B^ i 01
ii*i
                          Figure 6-1. Cluster arrangements.
                                        6-5

-------
     Many studies of plume dispersion from conventional  tall  round
stacks have been conducted.  The basic Gaussian  plume  model  (Turner,
     2
1970)  1s well  accepted for those uncomplicated  situations.   Unfortunately,
there has not been an adequate amount of study to thoroughly understand
the complex dispersion processes which affect entrapment of the effluent
plume Into the building and stack wakes.  However, several  modifications
to the basic Gaussian plume models can be made which approximate these
general adverse effects.   The modifications are  discussed below.
     Ground level concentrations due to individual gas turbine units
are calculated by using the Single Source (CRSTER) Model.3  The Single
Source Model calculates hourly concentrations for one year of hourly
meteorological  data.  Concentrations over averaging times longer than
one hour are simply the average of the values calculated during those
hours.  Calculations are made by employing the basic Gaussian plume
                                                          2
model and associated dispersion parameters (Turner, 1970).
     Two modifications to the estimation of plume rise in the Single
Source Model are made which result 1n generally  higher concentration
                                                                   A
estimates.  One, Briggs bouyant plume rise equations (Briggs, 1969)
are used to estimate the plume height at each receptor location.  The
use of the final plume height is avoided since maximum ground level
concentrations in most Instances occur close to the source.  This
modification is necessary since the final calculated plume height 1s
very high as a result of the extremely hot stack gas temperatures.  An
additional modification to the plume rise calculation 1s to use 70
percent of the Briggs estimate, as suggested by several recent studies
                                 6-6

-------
of stationary gas turbines (Hoult and O'Dea,  1975, Egan,  et.  al.,6
1975).  This accounts for reduced plume Hse resulting from strong
wind shear and Increased turbulence 1n the plume due to nearby
buildings.
     Dispersion of effluents emitted from short stacks 1s  affected
by increased turbulence induced by nearby buildings.  A recent study
(Huber and Snyder, 1976)  has suggested that the adverse effects  that
buildings can have on plume entrapment are best accounted for by
enhancing the vertical dispersion parameter.  Thus, the following
evaluation is made to determine 1f the building wake adversely
affects the effluent plume.  The plume center!1ne height at a  dis-
tance of 2 building heights downwind from the turbine structure 1s
estimated.  For those instances where the estimated height 1s  less
than 2.5 times the height of the nearby turbine structure, the
conventional vertical dispersion parameters are enhanced following
the suggestions of Huber and Snyder (1976).   This modification results
in higher estimated concentrations close to the source.
     Ground level concentrations are calculated with the modified
Single Source Model at receptors placed at 10 degree Intervals at
downward distances of 0.1, 0.2, 0.5, 2.0 and 20.0 kilometers.
Meteorological data for Dallas, Texas for the year 1964 are used
in the analyses.  Dallas, Texas Is selected as a site which 1s not
complicated by terrain and has mean wind speeds rather typical of
many moderately windy locations throughout the United States.
     In addition to the dispersion analyses for Individual gas tur-
bines, analyses of turbine unit clusters are conducted. The Single
                                6-7

-------
Source Model does not consider multiple source locations.   Thus,
the PTMTP (UNAM.flP) Model8 1s used for the cluster analyses.  The
model calculates hourly concentrations for up to twenty-five point
sources.  Again, the basic Gaussian plume model  1s used with Briggs
plume rise estimations.  Concentrations for a number of hours up
to 24 can be calculated, and an average concentration over any
nu.nber of hours can be additionally obtained.  Modifications to
plume rise and an enhancement of the vertical dispersion parameters
similar to those made to the Single Source Model are made to PTMTP
for these analyses.
     For the cluster arrangements, concentrations are estimated at
distances of 0.1, 0.2, 0.5, 2.0 and 20 kilometers from the downwind
edge of the cluster.
     Cluster analyses are conducted with PTMTP by using the meteoro-
logical conditions associated with the period for which the highest
ground level concentrations are estimated by the Single Source Model,
Also, the wind directions are aligned along the major axis of the
cluster arrangement.  This technique provides an Indication of the
maximum potential Impact.  However, 1t 1s somewhat limited 1n that
1t does not provide detail as to the variability of concentrations
throughout the year, as was possible with the Single Source Model.
     Plume rise from individual units 1n the cluster of turbines 1s
considered in the same manner as for the single source analysis.
However, there may be some reason to believe that interaction of
the plumes from the multiple sources results 1n enhanced plume rise
                                 6-8

-------
                        ft
(Millions et. a]., 7975).   Since It Is necessary here to determine
the maximum potential Impact, no enhancement Is considered since
maximum ground level concentrations occur on windy days where close
to the source building wake effects predominate over plume rise.
In addition, the highest ground level concentrations are expected
to occur close to the source before Interaction of many of the sources
may be significant.

6.1.4  Analysis of Air Quality Impact
     A summary of maximum concentrations found In the Individual
analysis of all nine gas turbines is presented below.  All short-
term maximum estimated concentrations occur at 0.1 km except for
the Solar (Centaur) turbine which has 24-hour max1mums at 0.2 km, and
for the G.E. MS7001B turbine which has 1-hour and 3-hour maxlmums at
0.5 km.  The output of the Single Source Model reports the first and
second maximum values which occur at each of the thirty six receptors
around each ring.  The number of occurrences of values which are 90%
and 50% of the maximum ground level concentration found among the
seventy-two values at the ring distance 1n which the maximum Is found,
is also considered.
     The highest annual concentrations occurred at 0.1 km except for
the Solar (Saturn) turbine which has maxlmums at 0.5 km, and for the
Solar (Centaur) turbine which has maxlmums at 0.2 km.  Since the usage
rate for most of the turbines 1s less than a full year, concentrations
lower than those estimated here, likely occur.  Annual averages
                                 6-9

-------
resulting from turbines operating 1n the spinning reserve  mode are
considered Insignificant.   Several  levels of the emission  rates
are used 1n the calculations for cluster arrangements.   For those
situations having all  turbines operating 1n the same mode, concentra-
tion estimates for other emission rates change proportionately.  Note
that under the spinning reserve mode of operation SOg emissions are
assumed to be one-third those under full load.
     It should be noted that 1n all cases NO  and N0? are  treated
                                            A       fc.
interchangeably for assessing the air quality Impact relative to NAAQS,
     The estimated air quality Impact of the Garrett A1research and
the Solar gas turbines presented 1n Tables 6-2 to 6-7 are  generally
small due to the small pollutant emissions of these sources.  For
almost all cases considered, even with 22 units operating  in a
cluster, it is estimated that no NAAQS are exceeded.  The  one
exception is the Centaur (Solar) model turbine.  With 22 units
operating in a cluster and using greater than 0.3/K sulfur  (S) fuel,
S02 NAAQS can be exceeded.
     Due to their size the General Electric turbines are charac-
terized by significantly greater emissions and air quality Impact
than the Garrett A1research or the Solar turbines.  Five different
units are considered below.
     The air quality impact of a single MS5001P unit operating 1n
either the regular mode or the spinning reserve mode is not estimated
to exceed any NAAQS (Table 6-8).  However, it is estimated that the
SOo 3-hour and 24-hour standards can be exceeded by a cluster with
                                 6-10

-------
                          Table 6  -2
       Summary  of Maximum Ground Level  Concentrations
for Garret Alresearch  Gas Turbine, Model 6TC8S-90  (GPUT-90)
1 ?J? Emissions
,j Fuel ) (g/sec)
Jet A Fuel
: 0.01 0.005
0.1 0.05
0 3 0.14
0.8 0.38
•
Number of Periods
within stated percent
jf maximum
90%
50%
CO Emissions
(ppm) (g/sec)
Jet A Fuel
321 0.4
90 0.11
,25 0.16
250 0.31
Number of Periods
within stated percent
of maximum
90%
50%
NO Emissions
(ppm) (g/sec)
Jet A Fuel
55 0.11
75 0.15
100 0.21
125 0.26


3-Hour S02 Average
(ng/m3)

1.8
17.8
49.7
134.0

9
55
1-Hour CO Average

0.2
0.1
0.2
1
71
Annual NO Average
(vig/m )

2.7
3.7
4.9
6.2

6-11
24-Hour S02 Average
(ng/m3)

1.2
12.4
34.8
94.4
i
1
9
8-Hour CO Average
(mg/m )

0.1
0.1 ,
1
29
Annual S02 Average
(wg/m3)

0.1
1.2
3.6
9.6









i
i
— Concentrations less than 0.1
i

-------
       o
       cr>
    oo {

    O"
    .Jj CO   I
    £5 
                   CM C •»*.  •  •
                 <->  rs CDCM r^
                 t/j      a.    CM
                      o —
                  J.  CVJ
                 OJ
                  
         E co •* in CO

         en co co ro co
         3.    co 01 in
        —          CM
                 CO
    ^-~   in
 to   o   o LO «* co
 c   o>   o o i— co
 o   co,—   ....
•r-  ^ QJ O O O O
 io   cnu
                  CM t—

                0
                     CO
            • o
            I r— CO CO
          O O O O
                                                                              £
                                                                •r- E CO r— r- CM
                                                             o c -^  •   •   .   .
                                                             o ro cno o o o
                                                              I
                                                             03
 en

 2
 *>  oo —
 > +JCO
** -i-  E CO i— i— CM

~          O O O
                                                                                0    ^^^-CO

                                                                                fl    0000
                                                                            O en
                                                                            

                                                                           3
8
                                                                                  0)
                                                                                     0000
                                                                                  --
                                                                                  ai i
                                                                      •  o in o
                                                               t: "o cvj  crt CM in
                                                               &    CO     r— CM
                                                               O-
                                                                                                                                       o
                                                                                                                                        c:
                                                                                                                                        (O
                                                                                                                                        co
                                                                                                                                        co
                                                                                                                                        O)
ra
                                                           
-------
                                           Table 6-4
                        Summary of Maximum Ground  Level  Concentrations
                            for Solar Gas Turbine, Model Saturn
   cflv Emissions

  (   . Fuel)  (g/sec)
     Na< >r&] Gas
  i.O            0.0
    Kerosene Fuel
  0.01           0.02
  u.l            0.17
  0.3            0.5
  0.8            1.33
Dumber of Periods
vithin stated percent
   maximum
         90%
         50%
3-Hour S(L Average
24-Hour S02 Average
Annual SO,, Average
       0.0

       2.3
      20.3
      59.7
     158.8
        0.0

        0.5
        4.6
       13.6
       35.9
       0.0
       0.3
       0.8
       2.0
                                  1
                                  4
   CO Emissions
 (ppm)        (g/sec)
   Natural Gas
   68.           0.27
   90.           0.36
 125.           0.49
 "-1.           1.0
    verosene Fuel
 163.           0.73
   90.           0.41
 125.           0.56
 250.           1.12
Number of Periods
within stated percent
of maximum
         90%
         50%
1-Hour CO Average
     (mg/m3)
       0.1
      "D.I
       0.2

       0.1
       0.1
       0.1
       0.2
 8-Hour CO Average
      (mg/m13)
        0.1

        0.1


        0.1
       8
      36
   NO  Emissions
 (ppm)        (g/sec)
      Natural Gas
   51.           0.33
   55.           0.36
   75.           0.49
 100.           0.66
   Kerosene  Fuel
   69.           0.51
   75.           0.55
 100            0.73'
 I01"-            0.92
Annual NO  Average
       0.5
       0.5
       0.2
       1.0

       0.8
       0.8
       1.1
       1.4
                                            6-13
                                                       —Concentrations  less  than  0.1

-------





















c
o
 4) 3E! CO
"* > CO
1  C _Q 4-"
t— 3 S_ C
.Q O 3 CO
<0 t-l— E
|— tO CO
in 01
E 
O 1- in
O 3
>»t- •—
i- 0


3
to
















t/1 - — .
+JCO
•r- E 0
ZD 010
"i
CO CM
n)
Ol
^
CM
0
CO
£_
3
O
3^
I tn '-^
V -t->ro
CM ••- e o
3 CDO
a.
LO 	










in ^~^
+JCO CD
•r- E •
rj 01
0) 3.
O) CM 	 •
(O CM
S-
0>
^
CM
O

s-
3

zc
1 CO- — -
CO +JCO
•^- E 0
:D OICD
"^
LO 	

*-" o
C CO *
O 10 CD •—

Ul Cjl CO 3
in ^^ "0 u.
•f™ ^-^
e CO
i5 ' — C
^-N n3 CO
CM r- S- to
Q m ^ O
00 3 -""• ° V
Li_ rO 'CO
i?I O ^
CO




CM LO «a- r*.
o us *a- to
i— oo LO r-»
CM to










co co csj en
CM en co «*•
i — m LO
•~











CTl O P- CM

r— CM r-^ 01
CO CM *a- r—
OO O-> LO
CM












CM *3- IO «d-
CT» CO O CO
^^ CO r""
CM tO

CM l~~.
O i — LO CO
... CO
O O O •
r—







O i— CO CO

o o o o



to

c E CM co «r co

01 o o o o
CM E
4) CM — •
Ol

*^
o
^J
^_

^
_o
-*• (^ ^. — N
^JCO
*~ •!- E i — C\J CM ^T
ZD Ol O O O O
E
LO-^
t^ tO CT*
"^ CM CO «d- O
V.S OOOr-
**^ VI
O O) fQ
*r™**—* C3
i/i
 ( —
*E *°
i3 ^
_j

g <0 0000

*2" OO O LO CD
S to en CM to
g-





^O (^O LO O"i
• • • •
o o o o











CM r— CM 00

CD O CD CD












CD
f^% CTl CO tO
i — Oi — CM












CO CM CM Uf)
CD CD O CD

r
C
J
•1
I
(
CO i— tO CM <
1 — *i- LO i — *•
CO O' O O i — c
3 I
U- <
•F
CO +
C 1
0) S
in +
£c
fl
CO t
i^ O O O O £
	 c
0-1 CD LO O C.
tO CTl CM LO
i — i— CM

fi-U

-------
                   Table 6 -6
Summary of Maximum Ground Level  Concentrations
     for Solar Gas Turbine, Model  Centaur
''" ,. Emissions
Fuel) (g/sec)
Neural Gas
0.0 0.0
\erosene Fuel
0.01 0.045
0.1 0.45
0.3 1.34
1 0.8 3.58

Dumber of Periods
i.'ithin stated percent
Df maximum
902
50%
CO Emissions
(ppm) (g/sec)
Natural Gas
68. 0.73
90. 0.97
125. 1.35
-°50. 2.7
Kerosene Fuel
126. 1.54
90. 1.1
125. 1.52
250. 3.05
Number of Periods
within stated percent
of maximum
90%
50%
NO Emissions
(ppm) (g/sec)
Natural Gas
89 1.5
55 0.9
75 1.3
100 1.8
Kerosene Fuel
86 1.7
75 1.5
'"'0 2.0
.<.5 2.5

3-Hour S02 Average
(pg/m3)
0.0

1.7
17.0
50.9
135.9



1
11
1-Hour CO Average
(mg/m3)
—
0.1
- •" 0-. 1
0.2

0.1
0.1
0.1
0.2



3
31
Annual NO Average
(pg/m3)
0.6
0.4
0.5
0.7

0.7
0.6
0.8
0.9
6-15
24-Hour S02 Average
(pg/m3)
0»0

0.3
3.0
9.1
24.4
I


2
11
8-Hour CO Average
(mg/m )
—
—
—
0.1

—
0.6
—
0.1



1
12
Annual S0? Average
(pg/m3)
0.0

—
0.2
0.5
1.4





























— Concentrations less than 0.1

-------



















10
c
o ^
5=7
. _> 1
£ .2m
^ 4-> LO
ft £ OJ
« 0 S-
c: -3
0 r- 01
0^-?
r— "° U-
r-~ S S: 0>
> o>
• « »SJ
us-1 £~
"O .,— +J
« Cj=> C
S o ^|
 C
E to 03
3 CD S_
E i_
•— s-tt
X 
1-1 'to
o s_ 3
- O i—
.>n ll (_J

fO
CO


1






-"

, p;
C "**^,
gj ZD en
0> CM-3
E ^
>
 cvj 3.
»O CM ^ 	 '
QJ
i^
^C

CM
O
CO
$K
3 (/)
O 4-> ^ — v
3— •^c'o
CO 3 -C.
CD
LO 3.
10 «J O
C OJ
O IO in O
LO "cntS
(/> . 	 .
•r- i —
E TO
U4 S-
CM r— -U
O tt> OS
CO 3 5T
LL_ 0
CO CD
S-

^1" CO O LO
....
LO CO CD I"-.
UD cr> o
r— LO








«* "d- O O
r^. ^f r^ ff

, —














LO LO O CD
CO (O CM I"""-*.
co co o r^
ro O vo
i— CM










*3* CM CD ^i"
^•~ ^^ f~ <"^
r-. CM CT>
CM in
LO
<*• LO LO 00
i — CD «3- **• LO
0) ....
3 O O •— CO
U-
O)
c
0)
I/)
o
at
O •— ro co
CD CD 0 CD


•i-CO
C El — CM CO LO

CnO O O O
CM E
O)
t.
0)
^

o
o
J_
o c/l^"
a: .t^^E i — <— i— CM
I C *^^, • • • •
00 3 aio O 0 O
E












£^~*
•r-CO
CnO O O i —
o>cv°~~'
•a
t.
ai
^
^^

o
o
i.
3 10
0 +>'-^
31 -r-ro
1 C E f™~ r~~ C\J CO
cno O O o

o t^ en ro r-
flj co
 in (O O O i — CM
C "«^ C3
O CD
VI rO
in t-
•i- 3
y— 1 ^
LU ^O
•z.
o o o o o
.-^ CO O LO O
E LO cn CM in
Q. r— CM
Gi-


ro CM CO LO
....
O O O O










I — r— i— CM

O O CD O














cn LO cn co
O O O r—












CM r*~ CM *^~
O CD CD CD

i— •* CM LO
O) LO i — LO O
13 ....
LL. i — i — i — CO

c
o
o
1
1
6-16

-------
                   Table 6 - Q            •
Summary of Maximum Ground Level Concentrations
for General Electric Gas Turblne.Model MS5001P
i SOp Emissions
v JS Fuel ) (g/sec)
DF-2 Fuel
0.1 3.93
0.3 11.8
0.5 19.67
0.8 31.46
Doped DF-2 Fuel
1.0 39.3
2.7 106.2
0.3 11.8
0.8 31.46
lumber of Periods
vitliin stated percent
)f maximum
90%
50%
CO Emissions
(ppm) (g/sec)
7.8 0.75
90.0 8.65
125.0 12.02
250.0 24.04 '
"•oped DF-2 Fuel
(.9% N)
5.3 0.56
90.0 8.65
125.0 12.02
250.0 24.04
500.0 43.1
Number of Periods
within stated percent
of maximum
90*
50%
NO Emissions
(ppm) (g/sec)
DF-2 Fuel
147.0 23.5
75.0 12.0
100.0 16.0
125.0 20.0
Doped DF-2 Fuel
(.9% N)
245.0 41.9
75.0 12.0
JO.O 16.0
125.0 20.0


3-Hour S02 Average
(yg/m3)

43.5
131.0
218.0
348.0

435.0
1175.0
131.0
384.0



3
27
1-Hour CO Average
(mg/m3)
—
—

_._. 	


—
0.1
0.1
0.3
1.3



6
53
Annual NO Average
(pg/m3)

2.5
1.3
1.7
2.1


4.4
1.3
1.7
2.1

6-17
24-Hour S02 Average
(yg/m3)

13.7
41.1
68.5
109.0
i
137.0
369.0
41.1
109.0



3
16
8-Hour CO Average
(mg/nT)
_ _
0.1
0.1
0.2


' (
0.1
0.1
0.2
0.9



2
13
Annual S02 Average
(wg/m3)

0.4
0.9
2.1
3.4

4.2
11.2
0.9
3.4






























Under Spinning Reserve Mode
for S0? emissions multiply by 0.33
ror 3-Hour SCL concentrations multiply by 0.67
or 24-Hour S02 concentrations multiply by 0.79
— Concentrations less than 0.1


-------
more than 8 units when fuel  greater than 0.5% S 1s used under the
regular mode of operation (Table 6-9).  The SOg Impact 1s less
under the sp1r..1ng reserve mode as a result of lower SOg emissions.
The CO average concentrations are well below NAAQS even for the
cluster arrangements.  Under the spinning reserve mode of operation
more occurrences of high concentrations are found due to the
greater CO emissions and the lower plume rise.
     The MS5002B unit is quite similar to the MS5001P, except it
has a shorter stack and does not operate under the spinning reserve
mode.  The maximum ground level concentrations caused by the MS5002B
are found to be 3 times greater than those caused by the MS5001P
(Tables 6-10 and 6-11).  However, at more distant receptors less
difference in ground level concentrations are found.
     The MS7001B units are the largest units considered 1n this
analysis.  Air quality Impact for an individual simple cycle unit
is estimated to be lower than NAAQS 1f less than 2% S fuel 1s used
(Tables 6-12).  Even with the reduced fuel rate taken into account
under the spinning reserve mode, the maximum SOp impact 1s slightly
greater than that found for the regular mode of operation.  In order
to meet the 3-hour S02 standard when a 16 unit cluster 1s 1n opera-
tion, 0.5% S fuel is necessary (Table 6-13).  The unflred combined
cycle MS7001B units are generally paired with their main stacks
set beside each other.  This configuration 1s assessed as two units
with one main stack.  The main stack diameter for this situation 1s
                                 6-18

-------
 I

CO
   -o
 oi  c
i—  3
XI  O
 10  i~
:xj  
 n3 C
 C3 (1)
     !CD
     •
     • i-
         O
Ct:
(/} •*""*»
CO O t\)O CSJ
co E . * . .
4J \ r- OOUD 
•r- m OO *d*CD 'Sj"
C! p. CM^" ^O
^D **—»•*
CO
OJ
0)
(0
^
O ^
-f
CM
s:
n~~ >=j- <*-o »d-
wE ....
+•> -v. CO COCO *3-
•r- CTJ p1— CM r— • CD
C 3 CMCO CO
3 —
CO
O£.
CO
ff—^
COCO CO 1 — O I —
1 \ f~ ....
f-' t
••— ~\ co O co co
c: cn r-~ coco i—
33. i — COCO "sf
0)
cn
£
0) CC
^? ^«.
"*• COCO O CO O CO
+o E ....
C3'NI-r- ^^. CO CO OO CO
00 3 3. CM I-.CM cn
3 ^^^ """ ""
0 "~
CO
'—• CO O O O
toco ....
•!-> E CO CM CM ^T
•i 	 ^. ^- CO CM CO
ccn CM r~ CM cn
Z3 3 r— r*-
CO""'


'•^^r— co r*—
o; co cn co co
CO • • • •
	 •<— CO CO O
r~
CO O
c at
•'" "'i-'aj s cn co t~- to
CO • 	 n_, 	 OO ,_ (;> r—
•r; r- r— co
CM i — O
0 
°fe
5" coco co r^ ^* co r-. ^t1
^•4^ E 1 . . . ...
^C.fw ^^ 1 , 1— | 	
C CT>
c_j-^ E
 r— r—
C E

00
Oi
CO — -
CO
co E CO
•r- O> r~I
C E


0)
cn
J-or
« ,-,
> COCO i— OOOi— " COOi—

_'r--^. O O r— CM Or- CM
O c cn

J.co^~"
^rj

CO
COE r— CO O r— COOi—

•r- CD O CD r— CM Oi — CM
C E
—3 s. 	 ^
CO




LO LO 1O »^ *~*
F*N. 
-------
                   Table 6  -  10
Summary of Maximum Ground Level  Concentrations
for General  Electric  Gas  Turbine, Model  MS5002B
S0» Emissions
(%S Fuel) (g/sec)
DF-2 Fuel
0.1 3.93
0.3 11.8
0.5 19.7
0.8 31.5
Doped DF-2 Fuel
i (.9% N)
! 1.0 39.3
,' 2.7 10.6
0.3 11.8
0.8 31.5
Number <>-; Periods
within stated percent
^f maximum
90%
50%
CO Emissions
(ppm) (g/sec)
DF-2 Fuel
7.8 0.75
90 8.65
125 12.0
25C 24.0
Doped DF-2 Fuel
5.3 0.56
90 8.65
125 12.0
250 24.0
Number of Periods
within stated percent
of maximum
90%
50%
NO Emissions
(ppm) (g/sec)
DF-2 Fuel
147 23.5
75 12.0
100 16.0
125 20.0
Doped DF-2 Fuel
245 41.9
75 12.0
100 16.0
U5 20.0


3-Hour S02 Average
(yg/m3)

139.0
418.0
697.0
1115.0


1393.0
3764.0
418.0
1115.0
-


2
21
1-Hour CO Average
(mg/m3)

-
, , 	 0.4
0.6
1.2

-
0.4
0.6
1.2



4
40
Annual NO Average
(yg/m3)

14.3
7.3
9.7
12.2

25.5
7.3
9.7
12.2

6-20
24-Hour SOp Average
(yg/m3)

47.3
142.0
237.0
379.0


473.0
1278.0
142.0
379.0



2
14
8-Hour CO Average
(mg/m )

_
0.3
0.4
0.7
t
_
0.3
0.4
0.7



1
6
Annual S(L Averaoe
(yg/m3)

2.4
8.2
1 1.9
19.0


23.8
64.7
8.2
19.0
































— Concentrations less than 0.1


-------
    O  OJ
 O
     « CD
. ,_ O) -r
   _Q  m
    S-  (U
   CD
       * 00 O 00

•r- \ en oo O oo
C COCO i— O r—
^ 3. CM r-. CM O\
•x^"- |— T—
oo
O)
(O
 CDpx. r— LO OO
3 CM OO LO
CO^-'
CD
ia
£-
•>
*t
to • — ^
—S^-nco en i--. o i^.
O .,_ £ • • • •
 c ^,00 r-. o i*.
, ID CD en *d~ LO en
J; 3- — - co h^ o f^
c: E LO en to rj-
13 \ LO LO OO LO
CDCM l~- CM O
CM 3. r- CM




V> (J CO
c o» en oo r-x LO
O LOr- ....
•— -~-x. a) co t— en f—
LO CD U ,— i— CO
to *^-*Ll_
*^

UJ |

CM i — O
O . o o LO
LO ^j- LO cn
OO OO CM LO
CM
LO CM i— i—

rv!r-^ CO 1-^
OOCM r»- o
CM LO r- 00
r~^ en CM LO
r~"





LO LO r»x r»-
oo 10 r*. r»«
oo o ^i* en
^J- i— CO LO
•SJ- CM i— CO
r™*


LO cv* r^ r-~-
LOl— LO LO
Locn r^ o
CMLO CM



. oo oo
r— CM O O



toco
4-> E i— CO OO f*--
•r— "*^ . . • •
c cno .— i— co
ra E

co

9)
o>
f tooo
«»••-> E r— CO •— CO
> -i- -C. . . . .
«t C cno O •— CM
CO x_x
°^"
3
^
00
to* —
4->OO
.C ~-x. LO LO CO
=> 0) ...
E O O i—
CM 	 	
toco
4-> E CM i— en en

c cno CM CM LO
3 E

CO*""
o
 vo O O
C \ •— CD CO CM ^
O 0> <*> •— «M
 LL.
1/1
^— CM


o
8
*-» co o oo
S. t->! d LO ca
o. en CM LO
^^ f-CM

•— co co r-.

O r— i— CO







r— CO r— CO
....
O O r— CM







. LO LO CO
...
1 0 0 •—


SM i— en en

O CM CM LO







i— oo co r~-
O r- i— OO







, CO O r—
1 • • •
1 O i— CM
r
C
J
J
«
r_ 1
3 LOLO •"
LL. LO LO O O (
CM'-^O CO CM «* ]
o%« •*
en I
x> • '
v^- *
Q_ *
0 <
^ i
coo o o £
LO o LO cr>
en CM to
	 --CM 	
                                                          6-21

-------
                           Table 6-12
        Summary of Maximum Ground Level Concentrations
for General  Electric Gas Turbine, Model  MS7001B  (Simple  Cycle)
S00 Emissions
(;.'S Fuel ) (g/sec)
Natural Gas
0.0 0.0
DF-2 Fuel'
0.3 26.5
1.0 88.2
2.7 238.0
0.1 8.8
0.8 70.6
lumber of Periods
within stated percent
3f maximum
90%
50%
CO Emissions
(ppm) (g/sec)
Natural Gas
2.3 0.39
90.0 15.3
125.0 21.2
250.0 42.4
DF-2 Fuel
90.0 15.7
125.0 21.8
250.0 43.6
(Spinning Reserve)
500 87.2
Number of Periods
within stated percent
of maximum
90%
50%
NO Emissions
(Ppm) (g/sec)
Natural Gas
119.0 33.4
55.0 15.4
75.0 21.1
100.0 28.1
DF-2 Fuel
163.0 46.7
75.0 21.5
100.0 28.7
125 0 35.8




3-Hour S02 Average
(yg/m3)

0.0

212.0
709.0
1906.0
71.0
565.0



1
6
1-Hour CO Average
(mg/m3)

—
_ ,*~~Q.2
0.3
0.5

0.2
0.3
0.5

2.4



4
27
Annual NO Average
(yg/m3)

0.4
0.2
0.2
0.3

0.5
0.2
0.3
0.4



6-22
24-Hour S02 Average
(yg/m3)

0.0

61.0
202.0
548.0
20.0
162.0



1
1
8-Hour CO Average
(mg/mj)

—
0.1
0.1
0.3

0.1 t
0.1
0.3

1.5



1
1
Annual SCL Avera^-
(yg/m3)

0.0

0.3
1.1
2.9
0.1
0.9































Under Spinning Reserve Mode
for S0? emissions multiply by 0.3'
for 3-Hour S02 concentrations multiply by O.uJ
for 24-uour S02 concentrations multiply by 0.98
— Concentrations less than 0.1


-------
   O).
   •o
 01

3


0)
o
0
0)
E
CO
^p-
CO 1
o  — • •* i— •— oo in
•r- en O CM 00 CM CO
C 3. CM
CD CO
O) r-
£
41
CM -— -
f) coco en co C3 en CM
•r- 	 O CMOOCMOr^
l_ C O) • i— Or- 1^ CO
3 :D 3. o CM r^ CM in
£ co~
i •—
CM
as
coco i— oo o r-» in
•>- -v. • in co in i— co
CO! O CO i— CO CM t*»
Z3 3. CM in r—
00
OL
CO
coco CM r^. O r*. co
*r~ <**vh . CO CO f^ fmm ^^
co> o oo i— co co en
Z3 a 1^ CO O CM O
— • CM r- CM
r—
Ol
CD
I
0~.3.cMCOr-
c33. ^»ooooinr>.
^ • — • h~ in en CM o
o co r^ co CM
i
CO
coco oo «* O cn in
••-••-«. • to cn en c— co
co) o co i— i— r- cn
=3 3 «O •— r-» CM CO
•— * CM in r-
oo
- — -co «* co cn in
oo co cn CD CM co
	 CM (•«•. CM
§ 3i !g o ^.in CM oco co
'« oi^ IScMcS^00^
,J2 *"*•— U. CM
^ _| 2!
55 2J^ cticoor^-r-oo
"- . v^o f— CM O O
^3 3E
C£.
W1CO CO
-t-> E
§\
<0
•r» "**>. • • . • . .
COl OOO OOO
fr
00
CO
coE co«±oo co^j-oo
•r-cn odd odd
00
• a:
1 CO CO
co E •
, i- O) i—
c E
2J co
^w E r*. cn cn r»» o cn
I5
i
CO
I/) E vo CO CO CO OO CO
T-O) OOr— OOf—
co
at
]•{ o> ai
.. *  1 c
uJ 
                                                                                                                     £
                                                                                                                     CO
                                                                                                                     u

-------
based on an equivalent double-unit circle.   The by-pass  stacks  are
treated Individually for this configuration.
     The MS70Gid unfired combined cycle unit 1s assessed as  a pair
of machines.  The reported concentrations  are, however,  given as  those
which occur per machine (Table 6-14).   Since the stack height is
significantly higher and the mass flow rate 1s very large for the
main stack, higher plume rise and thus lower concentrations  occur
near the source than for the other General  Electric units.  The maximum
1-hour and 3-hour average concentrations are found to occur  at  0.5 km.
It should be noted that concentrations occurring closer  to the  source
are primarily a result of the much smaller by-pass stack emissions.
As long as 2.7% S or greater fuel 1s not used, Impact of even a 16
unit cluster 1s estimated to be less than all NAAQS (Table 6-15).
     The highest ground level concentrations of all units examined
are found for the MS7001B (regenerative) unit.  Less than 0.8%  S fuel
1s necessary 1n order not to exceed short-term SOg air quality  stan-
dards at 0.1 km for an Individual unit (Table 6-16) under the regular
mode of operation.  Under the spinning reserve mode of operation,
maximum S0« concentrations are found to be slightly lower.  The only
situation 1n all the analyses of this report that CO concentration
standards are estimated to be exceeded 1s for the 8-hour average when
greater than 8 units are operating under the spinning reserve mode
(Table 6-17).  In order to meet the standards, 1t is estimated that
20% control 1s needed for 8 units and 50% control is needed for 16
units.  In order not to exceed the S02 short-term standards  at 0.1 km
                                6-24

-------
            Summary of Maximum Ground Level Concentrations
for General Electric Gas Turbine,  Model  MS700lB(Unf1red  Combined  Cycle)
,^0? Emissions
FueP (g/sec)
Natural Gas
0.0 0.0
DF-2 Fuel
0.3 26.5
1.0 88.2
2.7 238.0
0.1 8.8
0.8 70.6

lumber of Periods
vithin stated percent
)f maximum
90%
50%
CO Emissions
(ppm) (g/sec)
Natural Gas
2.3 0.39
90.0 15.3
125.0 21.2
?"\0 42.4
•^ DF-2 Fuel
90.0 15.7
125.0 21.8
43.6
lumber of Periods
within stated percent
of maximum
90%
50%
NOX Emissions
(ppm) (g/sec)
Natural Gas
119.0 33.4
55.0 15.4
75.0 21.1
100.0 28.1
DF-2 Fuel
163.0 46.7
75.0 21.5
100.0 28.7
J25.0 35.8


3-Hour S02 Average
(yg/m3)

0.0

36.3
121.0
326.0
12.1
96.7

.


1
5
1-Hour CO Average
(mg/m3)

—
, , 	 —
—
0.1

—
—
0.1



3
23
Annual NO Average
(yg/m3)

0.4
0.2
0.3
0.3

0.5
0.3
0.3
0.4

6-25
24-Hour SOg Average
(ug/m3)
•
0.0'

17.5
58.2
157.0
5.8
46.6
i



1
41
8-Hour CO Average
(mg/nr)
-
—
— -
—
—

— - c
—
~™™"


V
2
41
Annual S0« Average
(yg/m3)

0.0

2.1
7.1
19.1
0.7
5.7
































— Concentrations less than 0.1


-------







o>
i—
O
°
"O
01
C
gl
*0 l_ r—
fc. .— "^
g C IO
OCQ 3
«-> r— 0>
r— ° -r~
, 	 01 |5
• «£ s
•J (/)
01 C -Q 4J

(O S^ c
I— CD -5
... 
tE
t- v- CJ
i t»
3 O)
"* UJ .
(O
c:
(U
CD — >
i- CO
o »—
«*-
01
^£
.» QJ
cZ >
^* o>
01 M
2g
»— c
21 gr
o?o
OC. CO



O)
t.
O)
^s t/) *^~^«
"* -MCO
^ ---EO ^i-coor^o
S => cno oo r* co CM CM
vi 3. CM CM IO *3- ^J-
j_ (O — • ,— ^ r— CO
3 "~ "~
0
•^
CM










0)
O)
-^o co to o o t—
^03.0 u~> cr> CM LO o
S ' 	 r— •5J-COO'*
2 CO O 00 •— 00
.O i— CM
1
CO














t/1 (_> c\J
C  o> 3 CM op co r*«»
W ^— '<~ Ll- 4&CM
•r" fO ,*5y?
E s- CM
Ul 3 1
CM i — (O j» Q
O 
  • 
8
i-
2J
o
^c
oo





'





^K
2
01
<
o
CJ
t.
i
1


1










^^^
o
0)
I/I 
(A
M

I
o.

•




•i-ro
•r-E •— CMCO r— CM CO
=>O1 OC)CD OCDO
r—









t

, "-
tr~


V» ^
•i-CO
5\ ^U^r>; t1^^
o» o o o o o o
<£> E







r
C


<
J
I
<
<
o> *"
.... ... 1
in o 10 r— CM 10 f— co ,
o r~ *aj FT" CVI "* f
3
•— U.
fO • -«•
S- CM *
5 J.
arooooaooo 5
21 .... ... i
CM O LO O OLOO »
O» CM IO CM CM LO *•
i— CM ,- CM


6-26

    • -------
                              Tabie 6 -16
           Summary of Maximum Ground Level Concentrations
for General  Electric Gas  Turbine,  Model MS7001B  (Regenerative Cycle)
S0? Emissions
, v Fuel) (g/sec)
! OF-2 Fuel
0.1 6.4
0.3 19.3
0.5 22.2
1.0 64.4
2.7 174.0
0.8 51.5
lumber of Periods
within stated percent
Df maximum
90%
50%
CO Emissions
(ppm) (g/sec)
DF-2Fuel
5.0 0.97
90.0 17.5
125.0 21.2
2Fn.O 42.4
(Spinning Reserve)
500.0 84.8
Number of Periods
within stated percent
of maximum
90%
50%
NO Emissions
(ppm) (g/sec)
DF-2 Fuel
373.0 118.0
75.0 23.7
100.0 31.6
125.0 39.5

i





3-Hour S02 Average
(wg/m3)

160.0
180.0
790.0
1580.0
4280.0
1270.0



3
21
1-Hour CO Average
(mg/m3)

—
,. 	 0.6
'0.7
1.4

5.3



4
41
Annual NO Average

49.9
10.0 •
13.4
16.7






6-27
24-Hour S02 Average
(vg/m3)

55.0
166.0
277.0
550.0
1496.0
443.0



2
14
8-Hour CO Average
(mg/m )

—
0.4
0.5
0.9

3.5 t



1
7
Annual S02 Average
(pg/m3)

2.5
7.6
8.8
25.5
68.9
20.4

















1
i
I





I
Under Spinning Reserve Mode
for S02 emissions multiply hy 0.33
for 3-Hour S02 concentrations multiply by 0.63
for 24-Hour S02 concentrations multiply by 0.95
— Concentrations less than 0.1


    -------















^ 	
Ol
(J


CT ^
o *-
* S"7
£• — -to
CJ
c- CQ nj
L. ^^ ~
O CJ)
"oJoo 1Z
>z:

•"""OJ CO
"c o"""
3 ?-
^ " S
CJ 
at vi
01
S- C
•— c
3 f-
0) a
on oo

CE:
PO
irt CT
4-><
•r- Ol
C 3.
1^ ^— *
to
at
cn
 s:

CM 00 E
u 11
3
0 «5
1
rvi

^
CO

C 3.
— ^ «^_r.
co
C£
00-—
CO
oo E
-i-3 -•»,
•r— O^
— ) ^_J<

to


OJ
0>
2 21
<" "^oo^
< to E
cxi'i- a>
° 5^3
i/i
*- J2
^
ro

I/) CO
•>-> E
^) 3.
*-^-
oo


rv
00
'

^^.
00 (J
C OJ ^^,
O u>,_ 2;
>r" "^  3^
 >^-'i71
.^. ^~
-s ?
^"^U-
o '«Q^-,
U- oo
to yf

r-~ LO !•». CO O r—
cn co oo f**- i*^. ^^
CM oo *^ cn o PO
CM tO i— CM CM 00
r- CM LO r-







o i — o cn o r—
cn ti; LO cn co i —
r^* co cn co co PO
CM CO PO !-» LO CM
r— CM t — CM






to PO cn cn o CM
r— «3- r~s LO VO CM
vo co o r- to cr>
i— «* CO to PO CM


oo r-* o o o r*»
oo -d- «* oo oo oo
LO i — . cn oo cn to
LO to r-^ LO o *i-
i— CM LO ««d-
LO
f-^






*t CM CM CO O CM
i — • LO cn oS «5}* cn
CO P>. O r— r— ID
r— ^" OO ««3-
CM
CM




O CM O O O CM
r-^ cn LO o o *i-
co LO PO i — . to cn
LO r-. cn oo cn to
i — CM LO « E
'c "cn °°

s^<
to



0)
cn
 ....
c cn o i— CM «3-
N^^
3 to
o •—
1
oo

& — >
PO
00 E r— CM LO O
•4-* "^x. ....
•r- cn o i — i — PO
c^
00

to
coco
if s

i-O '
•


4)
CO^-
2 ct
£J U1PO CM OO ^1- 00
^ -r- V^ O CM PO tO
O c cn
o ^^5.
3^
J
^m

s_
OTO i — O ^f OO
•r- "^. O CM CM ^J-
c cn
,^^
00


oo
,3!
^ co
s-
*^ cn LO CM **• i/i
* o r-. i-^ CM a:
jg •" r— CM «*
0 0>^_ °^
*** 3 C
"i cti 'S.
8^Q oooo °
E LO O LO O 2
9- cn CM LO S
0. r- CM "^
                                                              o

                                                              c
                                                              to
                                                              0)
                                                              to

                                                              o
                                                              o
                                                               a
                                                              O)
                                                              u
6-28

    -------
when a group of 16 units are operating (regular mode)  In a cluster,
it 1s estimated that less than 0.2% S fuel  1s necessary (Table 6-17).
Concentrations found at 0.2 km and 0.5 km are one-half and one-fourth
respectively, of those found at 0.1 km for the operating 16 unit
cluster.  Thus, 1f the air quality region of concern 1s beyond these
further distances, less stringent control 1s needed.
     The results of the analysis (Tables 6-18 and 6-19) of the MS7001C
are nearly Identical to those for the MS7001B (Simple Cycle).   The
discussion for the MS7001B (Simple Cycle) 1s thus applicable.

6.1.5  Impact of Variations 1n Meteorological Conditions
     In general, the maximum estimated concentrations occur during the
windiest hours for most turbines considered.  Should windier situations
exist, even higher concentrations may occur.  In addition, concentra-
tions would be higher 1f situations exist when wind direction  variability
1s smaller.
     To provide an evaluation of how windier conditions effect maximum
ground level concentrations, further analyses have been conducted for
the MS7001B (Regenerative Cycle).  This unit 1s selected since 1t has
the worst air quality Impact.  Tables 6-20 to 6-22 contain the velocity
Increase evaluation for the worst case meteorological conditions with
only the wind speed modified.  Under the normal mode of operation, a
significant Increase 1n the maximum concentrations that occur at 0.1 km
1s found for the 1.5 times and 2.0 times wind speed cases.  Wind speeds
                                 6-29

    -------
                   Table 6-18
Summary of Maximum Ground Level  Concentrations
for General Electric Gas Turbine,  Model  MS7001C
S02 Emissions
(XS Fuel) (g/sec)
DF-2 Fuel
0.1 9.9
0.3 27.8
0.5 49.6
1.0 92.3
2.7 268.0
0.8 79.4

lumber of Periods
within stated percent
3f maximum
90%
50%
CO Emissions
(ppm) (g/sec)
DF-2 Fuel
2.1 0.51
90.0 21.9
125.0 30.4
250.0 60.7
(Spinning Reserve)
500.0 121.0
dumber of Periods
vi thin stated percent
)f maximum
90%
50%
NO Emissions
(ppm) (g/sec)
DF-2 Fuel
164.0 66.9
75.0 30.6
100.0 40.8
125.0 51.0


3-Hour SOg Average
(yg/m3)
62.6
175.0
313.0
583.0
1691.0
501.0

1
2
1-Hour CO Average
(mg/m3)
_ ,_J).2
0.3
0.6

2.7
4
17
Annual NO Average
(pg/m3)
0.2
0.1
0.1
0.2

6-30
24-Hour S02 Average
(pg/m3)
10.7
29.6
53.1
98.8
287.0
85.0
t
1
3
8-Hour CO Average
(mg/m )
0.1
0.1
0.2
t
1.7
1
3
Annual SO,, Average
(pg/m3)
0.1
0.2
0.3
0.9
0.3








Under Spinning Reserve Mode
for S0? emissions multiply by 0.33
for 3-HourSOp concentrations multiply by 0.8
for 8-Hour SCfL concentrations multiply by 1.6
— Concentrations less than 0.1

    -------
















o
U» i —
CO
00
T- ^^-^
^) J -.
Idgr-
£ .
•M i—
E 
 ,0 0)
i o> j_ a)
—1 3 V)
»0 1 	 '
a» c to -i->
t— 3 (O C
X> O U 
«•- in -P
O i- to
» C r—
i-  to
•o at
J}«
O)
J- C
«3 *-
r™ C
3 C
CD**-
a> a
a: oo




cc
(/I
t/)CO
J_J C
^^ c
•I- ->v.
c cn
r: p.
i U3
a. ^
cn
ia
*-x
O  — *
00

a:
to ^"^
oo
IA c:
•i->-^
•i- cn
C 3.
ZD — '

vo
"

ai
cn
m
£ z:
o a:
«t toco
*j (=
^•^ C™
CM'-- ^.
O c cn
in 33 p.
fc- vo""'
3 i —
o
£
1
CO

li^
oo
Irt c
-M-^.
•r- cn
c p.
Z3 — '
00


*****
ae.
»/\
w /



(/> U
C (U
O WT.O
•i- *^ ** 35
,_ __i --i S
K ^1^ —

1= «v
^CMPfe
o cu -r^
00 £ Si

« s
^™^



m cn «* i— o r—
21— CM r«. o oo
cn «* co m **•
r- CO VO 00 10



00 O CO O O CM
VO OO CO CM OO «*
co o oo tf cn cn
r— i— co cn CM




OO O VO. r— O VO
o in r~ CM cn CM
CO 00 in CO r- «(J-
r- CM CO CM





VO CO r— O O OO
i^ r>v t-s vo >«)• o
CM CO OO r— *1- CM
CM VO i— r— r- 00
r- CM VO i—









•* r>. O o o in
f-. CO VO r- CO CO
•— O co CM vo .
•— CM in r-





•


oo co vo cn o vo
o •* o in vo r-
co o o r«s vo «»
r— in cn vo co «i-
i— ^f r—



i— t>^
co CM in oo oo in
co o> vo o cn vo
i— CO 00 CM



oo
cn co vo co o >*
cn r^ cn CM oo cn
CM *r cn vo r»*
CM

r— oo in o r^ oo
O O O r- CM O


ee.
t/)*~*.
CO
 e
•4-* ^»
T 2*
=£
VO
P—

s.
in
Vr Ub>
< toco
0-^
0 co>
t_ ^S
W ^""^
3 VO
O •—
CO
S-
CO
iS^
ci1
— Jxl^
00
f^
oc.
to*-^.
CO
1 w E
1 +> -»^
I J
^"^
m
^U
^~


JM
2L
^z
? 2
!• •*•
« ^-^
> w>co
•r ** E
^ -r--^.
8C CT>
=1
€_ ^O
•• '•*
J'"


1

1 £~
CO
<« E
4J «*s^
^•^ ^^
•r* ^7)
=3.
00








a^
5S<
O cn TI
^^s
J2 £
w> **•
•MM
e csl
«S i
•MB
8_°
«?
|
Uk










. Oj CM ID
1 • » •
1 ooo





. r— CM <•
1 . • «
1 OOO

















inr*. «»•
...
1 OOr-










. 'd- vo CM
I * * *
1 f* .
01
(/)
c? *
cn CM
C i—'
gj

.r-
CX
f 4%
v>
o
•
8
in































,


















r-
o
c

^*
I/I
VI
0)

C/)
O
4J

    -------
              tj-
              VD
              01
              X
              CU

                 O)
               •> (/)
              to ro
              IO Ol

             r^ O
              (O C
           CU O i—i

      .—  <->•»>,
    Qi LL.  >, C 4J
    1  '   <_>  O -i-
    <_j
       CVJ
co
   ID
           (U
                 o
                 o
[Q -r- I-  CU
    •»-> 4-J >
    10 C
CO S- O) 4-
!•— cu o  o
O C C
O (U O  C
    cncj  o
CO CU    T-
~ Q£ i— 4->
       OJ  (O
       >  3
       O) i—
       _l  10
       3
       O

       CU
                                  CO

                                  $
                                 •Efo"
                                 I-  E     00 O COCO
                                                                                      to
                                                                                      •r— CO
                                                                                                      cvj
                     ~.
rvi
LO t-s CO r—
r^ co cvj
•st i—





o o«a- co
vo o r^ cvi
CO LO i—






o o ^-r*.
in cy> m r—
«* ^^ »V"

cu
o>
 cu
< to
<-, 1="™
O E Ol
"S^o
i. CVJZ
3 O «*
O LO II
i cvj cr
00



o"|
CVJ — •



cu
^* ^-^
£l
^t
t— 1 ^.



to
1-
•r-CO
!-.£=
in o>
• • .
r-O 0






00 CO r—
...
OO 0






(vj in cvj
• • •
r-O O
                o>
                3.    . -   -  _
               «—•   00 CO r—
                                                                                                                 O

                                                                                                                  C
                                                                                                                  (O
                                                                                                                                            to
                                                                                                                                            OJ
                                                                                                                                      
                                                                                                                                      C
                                                                                                                                      
(O

(U
                    CVJ
                   o
                   co co
                          3 O
                          o   •
                           I
                          CO
                                       to
                                       O  O)
                                                 OC3 O
                                                             CVI
                                                o
                                                CVJ
                                                      > ro
      u
      cu
      to
      —
      0>
                                 II
                                o-
                                  cu
                                  Ol
                                  10

                                  cu
         o o o eyi cvj
7" fc      	
h ^»    «* CD o oo *s-
"01    in ^- vo «et
= 3.    co in •—
   o
   cu
   to

E *CT)
                                                                                      to
         •r-CO
         H-  E

         O  O)

         cvj'3
                                                                                      Ol
                                                                                               O CO CO
                                                                                                 •   •  •
                                                                                               r-O O
                                                                                               cvi in cvi
                                                                                                 ...
                                                                                               i— O O
                                              s_
                                              3 O
                                              o in
                                              n: cvj
                                                                         u
                                                                        cr
                                       to

                                       i—s
                                      •r-CO
                                      H- E

                                      LO O>
                                          CD CD O r»> CT>

                                          O 00 CO 00 CVJ
                                          co vo oo in
                                          CT> CO CVJ
                                                           •r-CO
                                                           h- E

                                                           LO Ol
                                                        co co cvj r—
                                                          •   •   •   •
                                                        •— O O CD
                     O O
                                                                                                •  •   •   •   •
                                                                                                        CVJ O
                                                                                                           cvj
                                                                  6-32

    -------
                        
                                                                                 CM
                                                                                           tr» co ro in
                                                                                             •  •  •  •
                                                                                           r-.m CM o
                                                                                                                                            C
                                                                                                                                            (O
                        S-
                        Ol
                              u
                              cu
                              oo
    o
CM
    o
    UJ
 CU
    s
jJ5  LtJ
    UJ
    C3
              VO
              cn
              ns
              X
              CD
  oo •—- m
  co «O oo

 r^UD S
  CO     i_
 O  CU o

I  • •»  3 >_l
•  C  O1
I  O 'I- •>.
            CM
           O
           c_o co
                       1C i— O1

                       •*
                       CM
                             (U
                             to
                             COCO
                                              O O O O CO

                                              c\j in r«~ **• vo
                                              O» CM CO 00 CO
                                              r-- in «o CM
                                              CM •—
CM
 I   CU
_J_  >
O •!-
                     cu
       CO
       CO CU  u
       1- CU  o
       •«-> to J±
       E«—
       
                                                             o
                                                             o

                                                              s_
                                                              3
                                                              O


                                                             00
                                                                    o
                                                                    CD
                                                                    CD
                                                                 O *d-
                                                                 m ti
                                                                 CM cy
                                    •r—CO
                                    r- E
                                    in
                                        C7>
                                        3.
                                      o o o o •*

                                      «3- co i— r^. UD
                                      O 10 00 00 CM
                                      oo co co co
                                      LT) CO f—
  (U
  oo
   cu

O
oo co «a-
*!?• 3 O VO <
ol ° • "I 2
"r~ r—" rV
2 co
^
•^~
-(J
oo
III
LLJ

'cu
o
to ^"^
•«T ^v.
1— 0)
O~— '
CM
QJ
cy>cv"^
CU E
o o>
D E ;1



OO
CU
•^CO^
1- E

in cr>
• 3.
r— •*-'
o o o o en
O CM 00 CO «3-
VO CO O CM CO
0 0 0000
^J- CT» CO
r—

o o o o •—
O O* O i — VO
O«^_ *v i f*D m
r^ v NJ uu vu
CM r-. o r»>.
oo ^i- CM




oo o o en
o co r- vo in
CO VO CM  «*
i— CM Ol 00
CO OO CO


O)
Ol
03
i- 0
cu cu
> 00
*x. -^
CT
0 E
o o.*i-
(^ _
LJ. •
S- CM
3 O »*
o in ii
3: CM cy
I
r—







fll
lr
pf
o en
• E
CM* 	
*^~^
fQTO
CU E
O O)
0 EE
^» »~4Nw««



CO
cu
•r- CO
1— E

in o>
r—^-r
in CM CM r^


oo
VO





r-»
•
o
r—

r* coo

co 01 r** i~
«*T-OC





00 r—CO
• • •
10 coo


                    QL.
                "O  CU
                 E  O
                •r—  C
                 'S  CO
                 E  4J
                 3  oo

                S5
                                              •— CM m o o
                                                        CM O
                                                           CM
                                                                                           i— CM in o o
                                                                                             CM O
                                                                                                 CM
                                                                        6-33

    -------
                  CP>
CM

 I
                                         cu
                                           CO
                            O)
                            CO
                            re

                            cu
  CM
O 00
co
 ,  **
 t~ o
 3   •
 O i—
a:

   •—-   O O
 */)     re co       •   •
"^    CU    E
 CO    S-   —
in     o    en   co r-
  •  O C    3.
   Z HI	•
                       o o  **• «* «a-

                       r— co  in c*. CD
                       co o  vo i—
                       -*f CM
         co oo

8£3£°
                                              cu

                                              re
                                              s_
                                              CU

                                             <
                                                                         O
                                                                         OO
                                                                                     f-CO
                                                         CM
         CU
   O    «/)  .—<
   cu    re co


o. en    u   c*
0.000  C   E
                                                                                                CM LO LO
                                                                                               o CM LO •—
                                  CM
                                   II
                                  o-
CJ
(—
LU
_l
LU

|
^f
LU
^^
LU
CD
cu
__l CU "O
LU r- O
£^
o cu
CM >
1 CU S-
U_ > 
c
cu
o
c
o


f—
o
>
cu
_i

T3
C
3
O
i.
CD

E

E
•r-
X
re
ity Increase
o
o

'oi


o

c
o
•r-
4J
re
3
r—
re
>
LU



















CU

re
L.
CU
>
^c

CM
O 00

&^
S- O
3 •
O i—

1
co















u
cu
I/)

o>

LO
,
^_
CM
II
o-




                                         10
                                         (U
                                        LO
                                       o
                                       3:

                                       00
                         o •*
                         O 00
                         LO  II
                            o-
                 o     o o LO «?j- UD

                 •»v   CO LO CM 5t O
                  en   co en vo r—
                  3.   •* i—
                                                   I—   E    *f *O

                                                   in   en   co r^
                                                  LO r—

                                                  00
                                                                         c
                                                                         re
to
w
O)
                                                                                                                            to

                                                                                                                            o
                                                                         2
                                                0)
                                                U

                                                o
                                               U
             cu
             E   --
            •i-  co

                 •»,
            o
              •
            CM
                                             on
                                  i —  o
             0)
             I/)
             re
              •— CO
                 OO CO i—
                                                                         cu
                                                                         O)
                                                                         JO
                                                                         fc
                                                                         cu

                                                                         <.
                                                    U
                                                    OJ
                                                                                     •r-CO
                                                         CM
                                                                                      U
                                        cu
                                        E  —
                                        •i- CO
                                        in   o>
                                                   o o o
                                                               en
                      o Q. co    re co
                      O 0.00     0)   E

                      s- o »a-     o  "o>
                      3 000 O  C   E
                      o Ln  ii  z i-<  —
                      a:    o-
                       i
                                                          0>
                                           *•? ° 1>:'~l
                                           «* CM C3 C3
                  CO CTl VO CM
                   •   •   .   •
                  «a- •— o o
                                                   CM cr> r— r^ o
                                                   CM CM CO CM
                        O)
                        4J
                        re
                                                         H-  E

                                                         in  c^    «...
                                                           «  E    ** CM O O
                        10
                       LU
                        CU
                       -o
                        O
                               O)
                               o
                            2  
    -------
up to 25 m/s are considered 1n the 2.0 times cases.  It 1s Important
though to note that the set of worst-case meteorological conditions
selected as Input to the modified PTMTP Model are those which resulted
1n the maximum value among all the receptors considered.  The concentra-
tion calculations reported for the receptors downwind from the maximum
(0.1 km receptor) do not represent the potential Impact, since the
maximum ground level concentrations for each receptor will occur 1n
most cases under different meteorological conditions.  The worst
potential impact of a cluster arrangement at 2.0 km and greater
distances is best determined by assuming a simple multiplication of
the number of units times the maximum concentration calculated from
the Single Source Model analysis of the individual unit.  Under the
spinning reserve mode of operation, the increase in wind speed has
a much smaller Impact and 1n some cases results 1n a decrease 1n
maximum concentrations at 0.1 km.  Note that for both modes of opera-
tion the concentrations at the more distant receptor decreases as the
wind speed increases since plume rise has a relatively smaller change
and the dilution effects predominate.  These increased wind speeds
are possible occurrences; however, those at the upper end of the
range considered are not expected to occur very frequently.
     Under high winds, concentrations at about 0.1 km for the MS7001B
(regenerative) are estimated to exceed the NAAQS, even if less than
.3% S fuel is used.  The adverse impact from stationary gas turbines
is mainly a result of the low profile that the effluent stacks have.
A modest increase in the stack heights to about 2.0 times the height
                                6-35

    -------
of the highest surrounding structures  would  have  a  very  significant
Impact on air reality.   Tables 6-23 and 6-24 summarize the  Impact  of
stack height increase.   The MS7001B (Regenerative Cycle)  1s  again  used
as the demonstration case.  Significant decreases 1n  concentration,
especially at the closer receptors, are estimated to  occur.

6.1.6  Summary
     Nine gas turbine units are analyzed on  an  individual basis  and
in selected cluster arrangements.   Only one  situation is  estimated to
cause concentrations which exceed  the  CO NAAQS.   This situation  requires
that greater than 8 units be operating in the spinning reserve mode.
Many situations are Identified where the short-term S02  standards
could be exceeded.  Annual S02 and N02 average  concentration estimates
are not found to exceed air quality standards for individual units.
Although not specifically analyzed, annual average  concentrations
caused by cluster arrangements are not expected to  be large enough
to exceed standards.  Only the General Electric MS5002B  and MS7001B
(Regenerative Cycle) gas turbines  indicate any  potential  for exceeding
NAAQS.  if S02 emissions are controlled to meet short-term  NAAQS,  the
annual S02 concentrations will definitely be below  the standards.  For
the short-term averages the maximum impact of 16 units is approximately
5 times that occurring for an individual unit (Tables 6-10, 6-11,  and
6-16, 6-17).  If on an annual basis the factor  of 5 multiplication
holds, the maximum ground level concentrations  from 16 uncontrolled
units could potentially exceed the NAAQS for N02.  It is most likely
                                6-36

    -------
                s
                ro
                    CO
                to  to
                (O  «3
                (0  O
                o  c
CO
CM

 I

vo

 CO

J3
 03
QE:

o
UJ

UJ
CM O


U_  CO
         UJ  O)
       fff  I \

    O O J=
      .•i-  cn
10 co
           c .*
       ,-  
       I-  —   '
   O 0)  Q CO
       CO
       CO
         o  c

         co  o>r—  b

                >  c
                0)  o
                •o   -Q E in CO O CT> CTV
O \ Cn VOVOCNJr— O
o c  3. CM
z I-H 3: —
o
o>
to O
^ •
— '
 CO r- OTl CM O
«t t— cr*^ • — o cx> cr> i>. r^
l/> r— f—
CM o:
o
CO
i.
3 in
o
i V cJP
"^ JD E OOO^OO^
— o> co o in oo r»»
W 3. 00 CM ^f r-
a: v_- CM r-
(!) r-.
f) • »~ » o O O ^ •"
CO M E r— r-Oii— O
s_ .n — oo to *o in in
o :c en m in CM
o c -^. 3. i—
Z •— I U» ••— •
— • 3C
O
co »*• i. co
id • ja E
i-as-a-ir ^
coovo*^* CT> r^oc^^co
«£•— cyjc N— CM r— i— ey> o
r^ CM ** «*
CM i—
o
co
J- in
3 .
O i— ^-»
3: ii co
co z3 -^ °.°.0.*^
— o> o> CM «a- o cr>
w 3. 10 oo «(*• in co
n: ^ oo co i—
T3 CU
c o
•I- C «-s
2  E
c+J ^^ i— CM in o o
O •!- CM CD
O O CM


^^
CD O
«/» • ^^
(Or- CO
£ «« E OlCOr-
o c 5. i* ooo
^J ^ ^*» E
Zl—l IO 	
*-^ o:
0
CU
 in u v> E
< CM o":r — •
o
CJ
s- in
3 •
O r- •-»
OZ || CO
00 3?° •€ ">«^"-
^ 0, 000
I *-^
'~ CL<* II CO
 in n u> E o o
< CM o-ac ^-^
o
o
s- m
3 •
o •— ^
31 II CO
1 .O E 004- CM
•^ g) OOO
(/) E
•y» ^^^
i— CM in




CO
•
o



• •
00
• •
oo
•
o
O 0
CM O
CM
                                                                                                                                            c
                                                                                                                                            to
                                                                                                                                            in
                                                                                                                                            i/>
                                                                                                                                            0)
                                                                                                                                             in
                                                                                                                                             c
                                                                                                                                             O
                                                                                                                                       c:
                                                                                                                                       11
                                                                                                                                       o
                                                                                                                                       c
                                                                                                                                       o
                                                                                                                                       CJ
                                                               6-37

    -------
                 en
                  10
                  rO
                  X
                  a>
                  to co
                  fO tO
    O

<•  oc

~  t.
 I   UJ

CO  UJ

 (U  J
r—  "5

 ro  Ul

•-  £
    (0
              ^-N «0 O
               CO O C
            OJ -O    «-H
          C -M

          O J=
          •r-  O>
CO

O
o
r»
to
0)  t-  (O CO
>  CO  J- 3C
•r-  CO  -P
+J  CO  C J>i
 C
   to  
I/I
IO
4»
J_
u
o c
z_^"1



(U
moo

s- »*
 O
> •
CM
0
00
r—
3
C
C




CO
(/)
(C
O>
J-
o
o c
Z 1-4
>MWI^

(U
en to
 •
< r-
CM
O
to
3
O
z

«i-
CM



^"^
0)
to

o
o c
z «-<
^fc^



Jt/l

O
•
r—
II

X

to
DC
O
CO
10 O
Cn CM
in n
• .a
CM^
n to



if)
II
J2l
DC
to
DC

, —
t '
nJ3
U)
O DC
CO
CO O
cncM
CM ^
0-DC1"



in
•
fi—
||
.0
^^^
V)
a:
r—
*
II
==°
to
DC
O
Q)
CO O
"^x •
en CM
in n
CM -^



in
«
^—
n
(/)
DC






*-•»
CO
E

en
3.





, 	 „
co
E



ocoo
• •
CM CO
^J* r»









CO •«*•
• •
•
^^










CM
^ O Or-




CO
^
3.
*— '


CO^
^
3.



C'P
CD





^•^
CO
E
^^
3
^^"
^^
"'

o
o
i.
3
o
1








X

    -------
that for an annual average the maximum impact from 16 units will  be
somewhat less.
     An examination of meteorological conditions associated with
maximum concentrations Indicates that, 1n most cases, higher concentra-
tions can occur in areas with stronger and steadier winds.  Also, 1t
is found that concentrations can be considerably reduced by an Increase
of stack height.
     Discussion in the above sections reports the concentration estimates
made at the receptor of maximum concentration (.1 km for most cases).
The concentration estimates at the additional receptors are Included
in Appendix E (Tables E-l through E-26).  Because of the large volume
of data, the results in Appendix E are reported for only one set of
emission levels.  For the analyses of Individual turbines and the cluster
arrangements having all turbines operating in the same mode, concentration
estimates for other emission rates change proportionately.
6.2  WATER POLLUTION IMPACT
     Promulgation of Federal standards of performance for stationary gas
turbines will have little impact on water pollution.  Currently, the only
potential source of water pollution is from the water treatment system
for the water used to control NO  emissions.  Since manufacturers of gas
                                X
turbines predict that they will be able to meet NO  emission limits
                                                  A
within five years using dry control techniques  (combustor modifications),
there will be no impact on water pollution after dry techniques are applied.
     Relative consumption rates of water for alternative means of
generating power are shown in Table 4.4 and discussed in section 4.5.1.1.1
which show that the consumptive water use of a conventional steam boiler
                                 6-39

    -------
is over 20 times the water which might be used  to control  NO   emissions
                                                           J\


from a comparable gas turbine.   The boiler water and  the water used for



injection into the turbine both must be of high quality and the same



water treatment processes are used.  The boiler facility also uses  water



for other purposes such as cooling, blowdown,  fly-ash sluicing, wet



scrubbers, etc.  Therefore, the potential for  water pollution from  a



boiler is many times greater than that for an  equivalent turbine.



     A typical water purification system is shown in  Figure 4.15,   This



water treatment system for five 28 MW gas turbines operating  10 hours



per day treats 125,000 gal/day of water and rejects about  25,000 gallons



per day of waste water.  The 5,000 gallons used as backwash and rinse



water for the mixed bed demineralizer is treated in a neutralizing  tank



for pH adjustment prior to discharge to the municipal sewer,  return to



a river, or evaporation in a pond.  The 20,000 gallons per day of reject



water from the reverse osmosis system is sewered directly, returned to the



river or evaporated in ponds.  The quality of  the waste water prior to



sewering or return to the river is essentially the same as the influent



water except that the concentration of total dissolved solids in the waste



stream is about 3 to 4 times that of the influent.



6.3  SOLID WASTE DISPOSAL IMPACT



     Promulgation of Federal standards of performance will have little



impact on solid waste.  If dry control techniques (combustor  modifications)



are used to meet NO  regulations, no solid wastes exist.   The only
                   A


potential source of solid waste is from precipitation of solids in  the



wastes from the water treatment system for the control of  NO  .  If  these
                                                            A


are sewered or returned to the river, no solid wastes result.  If they



are taken to evaporation ponds, then solid wastes may have to be periodically
                                 6-40

    -------
removed and used as landfill.  Most companies contacted during the
process of developing emission limitations for stationary gas turbines
reported that they either sewered the wastes or returned them to the
river.  No quantitative data were available for a company using an
evaporation pond.
     As discussed in section 6.2, the water requirements for a gas turbine
using water injection for NO  control is less than 5 percent of the water
                            /\
requirement for a comparable steam boiler.  Therefore, solid waste, if
any, would be less than 5 percent of that from the boiler.
6.4  ENERGY IMPACT
     The energy impact of Federal standards of performance for stationary
gas turbines will be minor.  The only energy impact occurs from the use of
water or steam Injection for control of NO  emissions.  When combustor
                                          rt
modifications (dry controls) are used to reduce NO  emissions, there is no
                                                  J\
energy impact.
     The effects of water and steam injection on power augmentation and
gas turbine efficiency were summarized in Table 4.5 and discussed in section
4.5.1.7.  A combined cycle gas turbine installation using steam injection
suffers about a 1 percent decrease in overall efficiency at a water/fuel
injection ratio of 1.0.  A simple cycle gas turbine using water injection
suffers a 1 percent decrease in efficiency at a 1.0 water/fuel ratio.
However, if a small waste heat boiler is added to the simple cycle gas
turbine to provide steam for Injection purposes, then a net increase in
efficiency of about 1.0 percent occurs at a 1.0 water/fuel ratio.
     Table 6-25 shows the additional fuel consumption through 1980 if all
turbines sold in 1974 through 1980 used water in quantities equal to the
                                  6-41

    -------

8
+i
u
(O »~
i- 5
I-. CD

^\ 0V
Ol 0
Ol ^
LU
_f»
•?

|u
3 Ol C
4-> *-> O
CO «J •!-
Ol U
+* l"" QJ
*O 4J *•")
OC •,- C
«-•
,
z z: +•?$-
»-i i—OI
OQ O  a:
ui to a 3
z: =310 cc ii
tO OO •!-> 3
z oo«-i * •*->
Q O\\— Q) CO
o r— ir
3E Q> lO (O
O ^f > u> O)
f»4 F*^ ^C ^7 ^)
1^"" O^ ^^
•—• I— T3 r— — <
S
£r*
°*
cS
1— «








§
•f—

(J
^«
o.
— Q

eg
*
Ol








• o
U"
„
•~



CM

O













^_









O
o




^^
col
o
o
C3
CM





0»
O

M
r—


O

*o.
j£









(X)

••-•
. O







CO
in









o
in
CM
U3













r_









O
o
LO
in




















oo
CO















^
J3
•a
c
10
+J
to
O
m



o
o
CM

CO

f>.
en
at

•


o o

CO CO
r— r—












r— r—




•




O O
0 0
CM CM
M *
CO CO






§(^
o
o
* •*
oo oo





^
CO



o
en
r-.

U}
«^
CM





Ul
5- Of

2 ui ^_
o 
CX U) i. Ul
-«x 0) 3
r- r- .CXI
 -r- *-> C
c 0 Oi-
<1> ^^* ^•^
4^ lO J^
c






CM

CO









o

CO
r—












^.









O
O
CM
CO






f^t
O

CM










in
CO







Ul
§
•f«
+j
«
u
•r-
r-»
OL
a.
i.
UJ
JC

o
** 9
oo i-^
tx 00








CM ^f
co «r
• «
CO v-

01  H-
I U O
t— +•>•«- 1—
Xt 1 VI 421
1— cu eu il
§*ii.
«t-O^> *->
• uioa
C t— CU
Al in
jrf «J=0
IQ Ulr™*
4-» W*-r-
§JC JC

10 to
•a ui ui ui

    -------
fuel flow (1.0 water/fuel ratio) to meet a NO  regulation.  This high
                                             A
water rate and the use of water rather than steam biases the estimated
fuel consumption towards the high end since many turbines will use water-to-
fuel ratios of much less than 0.5 and, within the next five years, many
turbines will Incorporate dry control techniques which will obviate the
necessity to use water or steam Injection.  The data Indicate that even
at a 1.0 water-to-fuel ratio the increase in consumption of fuel oil by the
end of 1980 will vary between only 8100 barrels per day and 14,400 barrels per
day.  The estimated 1980 oil consumption of the United States is 21,000,000
barrels per day and the estimated total energy consumption of the United
States 1n 1980 1s equivalent to 47,835,990 barrels of oil per day.   When
compared to these figures, the additional consumption due to water Injection
varies between 0.04 to 0.07 percent of the total United States oil consumption
and between 0.02 and 0.03 percent of the total U. S. energy consumption.
     The data discussed in Chapters 3 and 4 for gas turbines burning low
Btu coal gas indicate that emission levels less than 75 ppm at 15 percent
oxygen can be met without water or steam Injection.  Therefore, as users
of gas turbines turn to coal gas fuels, the energy Impact discussed above
will be further decreased.
     The other energy impact of using water or steam Injection Is the Impact
of the power consumed by the water treatment system.  One user of gas
turbines reported that this power consumption was negligible when compared
to the power consumed by other auxiliaries at gas turbine sites.     One
source reported that the water treatment system consumes about 20 KM
                                        12
for each 1,000 gallons of treated water.    Using this number for the five
gas turbines producing 1400 MWH of power as discussed 1n 4.5.1.2, the
power required to operate the water treatment system is 1900 KWH or 0.01
percent of the power generated by the turbines..
                                  6-43

    -------
6.5  NOISE IMPACT



     The use of dry techniques  (combustor  modifications)  or wet techniques



(water 01  steam injection)  for  the control  of air pollution will  not



impact noise levels.  The stationary gas turbine  can,  unless  muffled,



constitute a sizeable point source for noise emissions.   The  application



of control technology will  not  influence these noise levels.



6.6  OTHER ENVIRONMENTAL CONCERNS



6.6.1  Irreversjble and Irretrievable Commitment  of Resources



     Application of wet or dry  control systems to stationary  gas turbines



will result in minimal or no short-term versus long-term trade-offs



between environmental parameters.  However, the application  of wet and dry



controls for NO  results in trade-offs between environmental  parameters and
               /\


energy and the selection of an  emission level results in impact on the



environment.



     As stated in section 6.4,  dry controls do not impact turbine efficiency



but  injection of water at a 1.0 water-to-fuel ratio decreases turbine



efficiency by 1 percent which is roughly  equivalent to a 3 percent increase



in fuel consumption.  This one  percent decrease in efficiency is higher



than that which will be experienced by most turbines since most turbines



will require less than a 0.5 water-to-fuel ratio to meet even a 75 ppm



NO   emission level.  As shown in Table 6-25, the impact on oil usage if all
  /\


turbines sold for the years 1974 through  1980 used water injection at a



1.0  water-to-fuel ratio is estimated as 8,100 to 14,400 barrels per day.



     Since almost any level of NO  emissions can be achieved using water
                                 X


injection by simply injecting more water at little additional cost, and



available data on dry controls  indicates  similar trends, the actual emission
                                   6-44

    -------
level foi  a regulation depends on factors other than the available control



technology.  Table 6-12 shows the estimated Impact of three different



NO  emission levels for all new gas turbines when a seven year period 1s
  A


considered.  This table shows that a 75 ppm NO  emission level will result
                                              ^


in decreasing the NO  emissions over a 7 year period by 650,000 to 980,000
                    J\


tons while an emission level of 125 ppm will decrease the 7 year NO
                                                                   o


emissions by 330,000 to 440,000 tons.  The table shows that at emission



levels of 100 ppm and 125 ppm, the NO  emissions are actually permitted to
                                     A


increase for turbines used 1n the "oil and gas", "private electric power



generation", and "other industrial" categories.  This occurs because



turbines In these categories now have uncontrolled emission levels lower



than the 100 and 125 ppm.



     Another alternative which could impact the environment Is the



implementation of a regulation which allows a delayed standard based on the



application and incorporation of dry controls on turbines.  If a seven year



delay were permitted for all turbines, Table"6^26 shows that the Impact



on the environment is the emission of 650,000 to 980,000 tons of NO ,
                                                                   A\


relative to a 75 ppm emission level.  If a 75 ppm emission level, based on



the use of water injection, were Implemented Immediately for large gas



turbines used in utilities with a seven year delay for smaller turbines



to incorporate dry controls, the Impact on the environment, as shown in



Table 6-26 would be the emission of an additional 38,000 to 68,000 tons of



NO , relative to an emission level of 75 ppm.
  J\


6.6.2.  Environmental Impact of Delayed Standards



     The environmental impact of delayed standards has been discussed in



section 6.6.1 where the impact of a delayed standard based on the application



of dry control techniques Instead of an immediate standard based on wet



or wet plus a percentage  of dry reduction was discussed.




                                  6-45

    -------
































CM
1
V*

_J
*"



















to
c
X
Vt
c
,2
•-• **
Eg
Q °
f £
So
U
C
I
S
t
u










«

1 !
S
!l

a|
Z tyi
2
v> o
(/I O
£^
ft
p
S5
*s
i
a
|
Q b









*
«>
s
I
E ~
s ** °°
in~lu
«C> Jl
d

3
1
4->
JP
d
w
t
&
CO
ac
iss
,_ """
in co
tx*o
«0
2«
g c
ES
1 '


won
Uncontroll
Emissions1
Lb/HWH

»7
1^1
« « 3
:r co

'Is
Is-?

•»_<>
» c
u*  N
o o


tn in


o o
8 8
in in



co co


~ 2
1 c
ot «
p> —



f 1



3
••

S 0
i i


» <••
O 0


o 5
I r
_
d d

O 0
0 0

S §
o o





I 1
O 0





§ §
m in



11



o o
» »
r- —
^ ^-



!i

T ^
t« **
s *
? t
ro fo
0 0
7 ?


to to
0 0
0 O


o o

5 5
o d

fM (M
3 3
0 0
to to
n ro
0 0
O 0





00 CO
S S
d d


in in


i i
in tn



II



11
in in
0 o



z 5

•5
t
I
t

7 R
d d


O tf
•- o


£ S
d d
So
to
*- d

oo in
at \o
0 0

00 IO
O 0

.c
5 J


S 2
— —
fi

in
1



















































,
I
S 5
S |
s C
5 S.
S °
1 5
"« i
¥»»—

*5 -I*
5f— M
e =•
** !fi
M C C
woo
S§ 5
If 1
aO ** w
*O Dtv 4J

•?«-««
" •« S v 5
§"• CT 1!
V ^J
jjj C^" ^ * m




(-46

    -------
                     REFERENCES FOR CHAPTER 6

1.  "National Ambient A1r Quality Standards"  published on April  30, 1971
    1n CFR 410, recodlfled to 40 CFR 50 on November 25, 1972.
2.  Turner, D. B.  "Workbook of Atmospheric Dispersion Estimates."
    Office of Air Programs, Pub.  No.  AP-26, Environmental Protection
    Agency, Research Triangle Park,  NC.  1970.
3.  Single Source (CRSTER) Model.  "Interim Users'  Guide to a  Computation
    Technique to Estimate Maximum 24-Hour Concentrations from  Single
    Sources (Draft)."  Environmental  Protection Agency, Research Triangle
    Park, NC.
4.  Brlggs, Gary A.   "Plume Rise."  USAEC Critical  Review Series TID-25075,
    National Technical Information Service, Springfield, VA.   1969.
5.  Hoult, D. P., S. R. O'Dea, G. L.  Touchton,  and R. J, Ketterer.
    "Turbulent Plume 1n a Turbulent Cross Flow:  Comparison of Wind
    Tunnel Tests with Field Observations."  Paper No. 75-49.1  presented
    at 68th Annual Meeting of the A1r Pollution Control Association.  1975.
6.  Egan, B. A., C.  Freudenthal, W.  G. Hoydysh, and A. Jepsen.  The ESEERCO
    Model for the Prediction of Plume Rise and  Dispersion from Gas Turbine
    Generators."  Paper No. 75-49.3 presented at 68th Annual  Meeting of
    the Air Pollution Control Association.  June 1975.
7.  Huber, A. H. and W. H. Snyder.  "Building Wake Effects on  Short Stack
    Effluents."  Prepared for publication in the preprint volume of the
    American Meteorological Society 3rd Symposium on Atmospheric Turbulence,
    Diffusion and Air Quality.  Raleigh, NC.   October 1976.
8.  PTMTP (UNAMAP) Model.  "Users' Guide to PTMTP," Environmental Protection
    Agency, Research Triangle Park,  NC.  1973.
                                6-47

    -------
 9.  Mllhous, M.  N.  F.  W.  Llpfert,  and  P.  Barontl.   "Multiple Plume
     Interactions from Rectangular  Gas  Turbine  Exhaust  Stacks."   Paper
     No. 75-49.2  presented at 68th  Annual  Meeting of the A1r Pollution
     Control  Association.   June 1975.
10.  Dupree,  W.  G.,  and West, J. A.  "United States  Energy Through the
     Year 2000".   U. S. Department  of the Interior.   1972. p  17.
11,  Letter from  Dunlop, D. D. (Florida Power and Light) to Goodwin, D.  R,
     (EPA).  December 1, 1975.  p 3.
12.  Letter and attachments from Assard, D.  G.  (Turbo Power and  Marine)  to
     Goodwin, D.  R.  (EPA).  November 26, 1975.  ,p.4.
                                    6-48

    -------
                            7,  ECONOMIC IMPACT1'

7.1 MODEL PLANT SELECTION
7.1.1 Introduction
      Section 3.1 discussed the basic structure of the gas turbine in-
dustry.  As was shown in this section, stationary gas turbines can be and
are used in a wide variety of sizes, use a variety of fuels,  are located
1n a wide range of climatic conditions, and are in operations which may
produce unique combinations of difficulties in attaining various levels of
NO  control.  To characterize better the range of potential impacts, a
  A
series of model plants which are typical of existing installations were
selected.  The selection of these model plants was based on factors consid-
ered fairly typical of all turbines in specific applications.
      The most important factors in the selection of the model plants were:
      (1)   capacity or size of the unit;
      (2)   heat rate characteristics;
      (3)   location in terms of the proximity of the units to an assured
            water supply; and
      (4)   operating hours.
The model plants parallel those applications discussed in Section 3.1.  The
applications selected for analysis were:
      (1)   standby power generation;
I/  All costs presented In this paper are stated in terms of 1975 dollars.
       *
                                   7-1

    -------
      (2)   oil and gas compressor stations;
      (3)   industrial internal power generation;
      (4)   utility p?ak load power generation; and
      (5)   offshort drilling platforms.
Table 7.1 shows the significant characteristics of each of these model
plants.
7.1.2 Description of Model Plants
7.1.2.1 Standby Power - Turbines in five size ranges were selected for
analysis as standby units.  The selection was based on data supplied by Gas
Turbine International* and are typical  of units likely to be used by hos-
pitals, computer facilities,  and the telephone industry when a continuous
supply of power is necessary.  Operating time for these units probably
averages around 80 hours/year, with a maximum of 200 hours occurring only
in the event of a major power failure.   Two hundred hours is considered
to be extremely high and not  likely to  occur very often.   It was Included
in order to bound the potential  range of short operating time applications.
        In the absence of environmental  control, the fuel consumption char-
 acteristics  (heat rate) of these machines  should be about 11,500 Btu's/hp-hr,
 or 15,500 Btu's/kilowatt hour for  a 1,100 hp machine using #2 grade dis-
 tillate oil.2
     •   The  large majority of these machines will be located  in  urban or  semi-
 urban  areas where adequate  supplies of drinking quality water  will be avail-
 able
 7.1.2.2  Pipeline Application — This application is, by  far,  the most diffi-
 cult to  characterize.  As indicated in Section 3.1, relatively slow growth
 is  expected  in  this  area, but the  potential  for economic impact is  great due
                                     -7-2

    -------
         o
                            I
                           o
                                          o
                                          in
                             o
                             in
                                      o
                                      in
                                                               o
                                                               oo
                           CM
                                                            10
                                                            ID
          O
               CO
                     o
                     o
                     in
      O
      o
      vn
        A
      in
                    o
                    o
                    CM
                      t»
                    co
                                          o
                                          o
                                          CM
                                           *
                                          CO
                                      o
                                      o
                                      CM
                                        A

                                      CO
                     O
                     O
                     CO
€/>
o
 o

CM
      CM

      «**i
                     O


                    CM
                                          O


                                          CM
                                       O


                                      CM
                                                                         6

                                                                        CM
 C     *•

-  "  S
4J  l_ iJ

 <0  3 ^
 t-  O
 0> OC
 o.
o    o
o    o
CM    CM
 I.     I.
O    O
CO    CO
                                          o
                                          o
                                          o
                                          OO
                                          o
                                          o
                                          CM
o
o
o
CO
                                       o
                                       o
                                       o
                                       CO
                                                   o
                                                   o
                                                   o
                                                   CO


                                                    I


                                                   8
                                                   in
                co
                N
               «r~
               to
 o.    o.
JC    JC

o    o
in    o
CO    f-
                    o
                    o
                    o
                    CO
                                       o
                                       o
                                       o
                                       CO
                                                               o
                                                               o
                                                               o
                                                               vo
                                                               to
                o
                10
                o
                o
                                          0)
                                          4J
                                                o
 c     c
 (O     IO
JQ    .0
 »-     $-
=>    H>
                              C    
                              O    t.
                              •r-    O
                                                                     •r-  C

                                                                     4J  O
                                 C    Q- (O

                                 It)    E -M
                                 r—    O «/>

                                 Q.    O       O
                                    CO
                                    «*-
                                                                                               «*


                                                                                                O
                10
                u
                      I
                      o
                      CL
                      T3



                      I

                      00
                     eo


                    a.
                •r-   C

                S-   t-
                4->   Ol

                Vt   +J

                3   C
                                    co
                                   •r- M-
                           QL
                          •^«

                          D.
                                                                          •o


                                                                          8-
                                                                          o
                                                                                       C7)

                                                                                      i
                                                   .  7-3

    -------
to potential competition from reciprocating engines.   Typical  pipeline
applications can geographically range from above the  Arctic Circle to the
Monave Desert.  Onsite water may or may not be available, although one major
pipeline company in che Southwest has indicated that  no  present  or planned  pipe-
                               3
line is without a water supply.
       The machine selected for analysis is a 4,000 hp turbine which may be
used in groups of two or three, operating 8,000 hours/year.  This particu-
lar turbine is of relatively small size, but is commonly used  in pipelines.
Also, if the imposition of NO  controls were to have  an economic impact, it
should show up on the  smaller turbines which cannot take advantage of the
economies of scale (lower  cost  per gallon)  of  the water  treatment  facilities.
Water is assumed to be available onsite or available  within fifty miles by
transport.  The importance of water availability will be discussed in the
next section.
       Fuel consumption in this application is very dependent on fuel type.
Rolls-Royce, which has made more  inroads into American markets than any
other foreign manufacturer, has recently announced a  new turbine with a non-
                                                                       4
regenerative fuel consumption rate of 7,200 Btu's/hp-hr on natural gas.
The turbine made by the major U.S. manufacturer, Solar, has a fuel consumption
of 9,750 Btu's/hp-hr, or  13,200 Btu's/kWh, and this rate is not very dependent
on fuel used.5
       If water is available, the turbines selected for pipeline applications
are also typical of industrial applications which will act as caseload
machine drive or electrical generating applications.
7.1.2.3 Electrical utility applications - At present, the two chief char-
acteristics of turbines used in the electric utility  industry are large
size and relatively low running time (500-2,000 hr/yr.).  Future applications
                                     7-4

    -------
of turbines could, 1f combined cycles become more frequent, Increase 1n
operating time up to medium usage or caseload applications (4,000-6,000
hr/yr.).  As discussed In Section 3.1, this application may not become as
prominent as once expected due to the high cost of oil and the reduced
availabilities of natural gas.
       For purposes of this analysis, a 66 MW simple cycle turbine operating
500 to 2,000 hr/yr on #2 distillate fuel was selected.  This particular
machine is about middle  size on the list of machines built by major
manufacturers.  Also considered was a baseload application (8,000 hr/yr)
which might occur under a major breakdown of baseload generating capacity.
The turbine selected is designed to produce 66 MW at baseload, but peaks for
short time periods at 72 MW.
  \
       For purposes of this analysis, a combined cycle system was not con-
sidered, because it was felt that the water quality system would.be the same
for combined and simple cycle machines and the Increased capital cost of
the combined cycle would only serve to make any Impact smaller.
       All three utility applications were assumed to be located where
adequate supplies of potable water were available.
7.1.2.2.2 Offshore applications -One special application has been broken
out for study.  Turbines are frequently used on offshore drilling well
platforms (up to 21 per platform)   for various mechanical drives, gas and
oil pumping and electrical power production.
       The turbine selected for application was a 4,000 hp machine operating
8,000 hr/yr.  Fuel for this application was assumed to be natural gas with
a fuel efficiency of 9,000 Btu's/hp-hr.
       The primary reason for breaking out this relatively small market for
turbines was the potential impact of requiring sea water as feed to water
purification equipment.
                                    7-5

    -------
       In summary, six model plants based on common industry applications,



considering location, size and.fuel mix, were selected.   (See Table 7.1)



These data will be used in the following sections to determine the economic



impact of the proposed standards on the gas turbine industry.



7.2    COST ANALYSIS OF ALTERNATIVE EMISSION CONTROL SYSTEMS



7.2.1  Cost Analysis of NO  Control for New Turbines
                          A~  "^««"^^^""^^  —^^»™B«»^B™™^^^^™.«»»


       Thermal NO , as discussed in Chapter 4, can be controlled in several
                 A


ways:  wet control (water or steam injection), combustor modifications,



catalytic cleaning of the tail gas, or some combination of the three



techniques.  The most prevalent control technique practiced by turbines



meeting a stringent state or local requirement is a combination of dry



controls plus wet control to bring the emissions down to a 75 ppm level.



The emission rate of NO  without water injection will vary from manufacturer
                       A


to manufacturer depending on the success and application of dry control



techniques.  Table 7.2 shows estimated NO  emission rates for various £ur-
                                         A


bines offered for sale in the United States.



       All of the data in Table 7.2 relate to simple cycle machines operating



a baseload conditions.  Although data are not complete, it appears that re-



generative turbines will have higher NO  emission rates than simple cycle
                                       A


machinies over the entire range of operating conditions.



7.2.1.1 Water  Injection - Both the amount and quality of water injected into



any turbine are a function  of turbine design and will vary from manufacturer



to "lanufacturer and from model to model produced by a company.  Table.7.3



lists the quality of water  required by the various turbine manufacturers.
                                    7-6

    -------
                                 TABLE 7.2

                 EMISSION RATES FOR NON-REGENERATED TURBINES
Manufacturer
Solar
Ref.


 8
 Size
1100 hp
3830 hp
Emission Rate
  at Base
       lSX 0)
   60  - 81
   90  - 153
6M
               3000 hp - 4000 hp
                     125
Westinghouse
 10
 66 MU
     219
Turbo-Power
 Machine
 10
                     66 MU
                     242
General
 Electric
 Turbodyne
 11
 12
    MW
 60 MW
    163
                                                                       340
                                      7-7

    -------
                                 TABLE 7.3

                       WATER QUALITY SPECIFICATIONS
Total Dissolved
 Solids (TDS)
      +

Non-Dissolved

 Solids (ppm)
Sodium
  +

Potassium (ppm)



Silica (ppm)



Particle Size
                        Turbine
1.0 - 5.0
                          0.5*
                          0.02
                    Boiler Feed
0.25
                       7.0 - 8.5
                         0.25





                         0.0




                    Not Soecified



                      6.5 - 7.0
                 Electronics
                                                                    0.1
               Non-Detectable
                     6.6
* Turbo Power and Machine limits sodium to 0.1 ppm.
                                    7-8

    -------
 It is  not  clear  exactly what  quality  of water  is  required;  the manu-
facturers'  estimates allow for a significant margin of safety.   The
last column of Table 7.3 shows the quality of water used by a West
Coast electronics firm.14yf   Water of this quality is widely used
in many applications such as high pressure steam boilers, electronics
 food and Pharmaceuticals.  The  technology  is available  to achieve  this
 quality level; however, it will  not be required for  supplying feed-
water for wet control systems.  One of the primary costs of NO  control  for
turbines is the preparation of water to meet the strict quality specifications.
Table 7.4 shows the estimated quantities of water required by the model
plants to meet the 75 ppm, 100 ppm and 125 ppm levels.  It is Important
to note that the only model plant which may require additional  control
to meet the 125 ppm level is the utility application.
       The costs estimated in the next sections are based on meeting the
75 ppm level burning liquid fuel.  Gaseous fuels typically will result in
lower NOY emission levels and, therefore, will require less control at a lower
        A
cost.  Section 7.2.1.1.7  discusses the change in control cost 1f the desired
NO  level were 100 or 125 ppm.
  A
       A wide variety of equipment can be used to deliver water of the purity
desired.  Two main components of a purification system such as this are
reverse osmosis and de-ionization.  Each process has been described in
Chapter 4.
7.2.1.1.1  System selection and cost for model plants - In system selection
and sizing, a specific design philosophy was followed'.  Based on EPA's policy
in such areas as flue gas desulfurization and other control equipment
where system design and operational capacity are sized for normal operating
practice based on engineering judgement, large amounts of redundancy were
                                    7-9

    -------
                                  TABLE 7.4
              ASSUMED FLOW RATES  (gpm) FOR ALTERNATIVE STANDARDS
Turbine           75 ppm              100 ppm                  125 ppm
 Size           Standard             Standard                Standard
                                        0                       0
                                        0*                      0
                                       35                      25
1100 hp
4000 hp
66 MW
15
100
42
*As,sumes that further limited dry control of the GM [nachine 1s possible.
                                       7-10

    -------
not built into the water purification system.  If, for example, a turbine
required 60 gallons per minute (gpm) of capacity, a system capable of main-
taining 60 gpm would be considered adequate; significant amounts of water
storage would not be needed.  Alternately, if normal operation of fourteen
hours per day required 50,900 gallons per day, a system capable of pro-
ducing a flow of 35 gpm plus a storage tank of about 20,000 gallons would
probably be adeqaute.  A similar situation exists in regard to standby
units.  In any normally occurring operational pattern, these units would
not be expected to run more than a few hours at a stretch.  Tap water,
which can be upgraded using existing water purification equipment, would
almost always be available.  The circumstances in which these units would
run for several days and the water supply would be disrupted, as in the
event of a major power failure (perhaps lasting longer than one week), need
not be specifically designed for.
       Redundancy is often built into the water treatment system to avoid
turbine shut down during water treatment system outages.  This concern for
system reliability does not appear justified.  A turbine designed for water
injection can operate if the-water is .not injected into the combustion
chamber.  Also, the systems required to purify the water for injection are
well proven and reliable.  San Diego Gas and Electric has reported that in
the last one and one-half years of operation, the water system has out-per-
formed the turbine in terms of system reliability.
       In this analysis, no effort was made to design'an optimal or least
cost system.  Rather, what was felt to be a workable system with adequate
capacity was selected.  For example, in some situations a water system
                                    7-11

    -------
designed for continuous application, coupled with a storage tank, could
prove to be the most economical.   This will  vary from site to site and was
not considered.  Thus, the systems selected  are probably less than optimal
from a cost standpoint, but adequate to meet the requirements for water
quality and water injection rates.
       An exception to the above was made in the case of electric utility
operation, where actual installed costs for  installations using water-
injection systems were used.  It is assumed  that, in this case, an effort
                               i
was made to optimize the overall  design.
       Special cases of clean water availability or non-availability were
also not explicitly considered.  An example  of clean water availability
might be at a utility which has a large amount of built-in redundancy in
its boiler feedwater makeup system.  If it is possible to use some of this
redundancy to supply water to the gas turbine, the incremental turbine
control cost would be very low.
       A similar series of decision rules are established to select cost data
for use in the cost analysis.  For each water supply rate required, a series
of equipment manufacturers were contacted.  Each manufacturer was asked to
suggest not only equipment and costs for each flow rate, but also to identify
how equipment costs could vary with incoming water supply.  These cost esti-
mates were then checked against estimates obtained from users and manufacturers
of gas turbines.  Whenever possible, any system which appeared consistent with
all estimates was used in the model plant analysis.  These data were compared
with water supply systems actually in use to control NO  emissions from gas
                                                       /\
turbines.  In all cases, the data used in this analysis are consistent with
actual operating experience with water injection NO  control systems.
      •                                             rt
                                    7-12

    -------
7.2.1.1.2 Standby power - The first model plant type selected for study was
on emergency power standby units.  The two unit sizes selected were a 350
and a 1,100 hp unit.  Based on manufacturers1 recommendations,   15 gallons
per hour was selected for use in the larger unit; in the smaller unit use
was scaled down to 6 gallons per hour.  However, the possibility exists that
these units may not require water injection to meet either standard (See
Table7-2)  The low expected operating hours of these turbines have led two
             17 18
manufacturers  '   of water purification systems to recommend that portable
units be used.  The portable system, in one case, consists of a nine-inch
diameter tank containing a cation exchange, an anion exchange, and a mixed-
                                           19
bed system.  According to the manufacturer,   this system will provide water
of the required quality at a purchase cost of about $1,200, including con-
trols.  The installation cost of the system is less than 15 percent of pur-
chase price.  The system should provide about 2000-4000 gallons of water
between regenerations in urban areas where United States Public Health
Service (USPHS) water quality limits are met.  Waters which are higher in
dissolved solids would result in shorter intervals between regenerations.
Cost per regeneration, about $60-$70, would be incurred probably not more
than once per year.*  Water from such a system would easily exceed all manu-
facturers' specifications for quality.  Other manufacturers offer similar
equipment at both higher and lower costs.  One offers a system which, assum-
ing an input water loading of 300 ppm TDS, would provide about 500-600 gallons
of capacity (or three months normal operation) at a purchase price of $260
                                20
with a regeneration cost of $33.    This equipment is laboratory based
and may or may not be suitable for industrial applications.  The system
selected for application to the model plant was the tri-bed system costing
*  If the system provided 3,000 gallons between regenerations, a turbine
   requiring 15 gallons per hour could be operated 200 hours per year.
   For this system, the operating cost would be $20/1,000 gallons.
                                     7-13

    -------
$1,200 plus $200 for installation such as piping,  water inlet,  etc.   An
annual operating cost of $60, or $20/1,000 gal,  was established based on  an
estimated one regeneration per year; i.e., if the  larger of the two  turbines
was operated 200 hours per year.
Model Mant
Standby
350
1100
7,2.1.1.3 Basel
Capital
of Water
Total

$1 ,200
1,200
oad turbine
Cost
Treatment
$/kW
i
$4.68
1.50
for industrial
Operating Inst
Cost
$/1000 gal $/yr
$20.00 $9.00
20.00 22.50
application - The model plant
allation
Cost

$200.00
200.00
for
this application was a 4,000 hp (3,000 kWh) machine, for which operating time
was 8,000 hr/yr.
                                     21
       According to one manufacturer,   water injection requirements for this
size unit are about 100 gal/hr or 1.67 gal/min.  Two different systems were
selected for this machine based on two input water qualities.   A water very
high in TDA, iron and manganese was assumed to be treated by a system con-
sisting of a complete pretreatment operation, including an aeration system,
a clarifier and an activated carbon system.  This pretreatment was followed
by * R.O. unit and two mixed bed de-ionizers, because one unit must be on-
line while the other is being regenerated.  Alternately, a storage tank could
be built and only one DI unit used.  Capital costs for this system ranged
from $8,000 to $25,00022' 23' 24' 25' 26 based on manufacturers' estimates.
                                    7-14

    -------
Most estimates for the system were from $10-$12,000.  Most manufacturers
felt that since this system was smaller than normally built for other
applications, unit costs might be high.  Data from one turbine user showed
                                                                  ?5
an estimated capital cost based on this type of system at $7,200*.
       Installation costs for equipment such as this, from $400 to $2,000,
are relatively low because the equipment is skid-mounted.  Depending on the
manufacturer, the necessary equipment would occupy from 10 to 30 square feet.
       Operating costs for the above equipment will vary from location to
location depending on the requirements for regeneration, the cost of re-
generation chemicals, and power for operating the R.O.  This cost 1s esti-
mated by one turbine user as $1-2/1,000 gal of water produced, or $864 -
            28
$l,600/year.    San Diego Gas and Electric, although operating a much larger
system, estimates that the cost will be in this same range of $1-2/1,000
    on
gal.    Manufacturers' estimates were in the same range; one quoted $4.00/
1,000 gal or $3,500/year.    Labor costs are not Included in the estimates
but are expected to be minimal.  No more than two hours of labor per week
will be required.  In unmanned stations, no more than one visit per week
might be needed and it coincided with normal site visits.
       Sufficient supplies of water may not be available onsite at pipeline
or other remote units.  Figure 7.1 shows the cost of transporting bulk q'uanti-
                                    31
ties of water over varying distance,   including all costs associated with
*  This number is based on a capacity of 2,400 gal/day for the system
   (1.67 gal/min x 1,440 min/day) and an estimated capital cost of
   $3/gal of daily capacity.
                                    7-15

    -------
          o
          ZJ
          k.
         O
         O
         O
                                                                             in
                                                                             cxl
O
O
O
                                                                             o
                                                                             o
                                                                             in
                                                                              o
                                                                              in
                                                                              in
                                                                             •cvi
                                              7-16

    -------
water delivery except unusual detention of the truck at the delivery site.
For this analysis, 1t was assumed that treatment would be done at the site
and that a storage tank of 14 day capacity would be provided at an additional
capital cost of $18,000.
Model Plant
                    Capital Cost
                  of Water Treatment
                  Operating
                     Cost
                   Installation
                        Cost
Pipeline
    4,000 hp
                  Total
                 $12,000
                               $/kW
$4.00
              $71000 gal  $/yr
$3.00   $1,750.00     $2,000.00
                                                                             32
7.2.1.1.4 Offshore platform - If the same type of turbine were installed
in an offshore drilling well platform, the type of equipment needed to
provide a supply of water is quite different from an application where
potable water is available.  Information provided by one turbine manufacturer
estimted the capital and installed cost of distilling sea water would be
about $18,000 ($24/1,000 gal).   Exhaust heat from the turbine would be
used in the distillation process.  Automatically vented to the atmosphere,
the heat would not represent an issuing penalty.  The distilled water
would be further treated with a deionizer before being injected into the
turbine.  Since the quality of distilled water will be higher than normal
drinking water, the size of the de-ionizers can be smaller, probably cost-
ing in the vicinity of $3,000-$4,000.    The costs for operating this type
                                                                  o *
of system are expected to be  in the range of $11-12/1,000 gallons.    One
turbine user,   operating a 22 MW regenerative turbine (7 times the size of
the model plant and requiring 8 times the water for injection), estimated a
                                    7-17

    -------
capital cost of $82,000 or $4/gal/day of capacity.   This cost,  which did
 i
not include a de-ionizer which may very likely be required,  was substantially
lower than the $9/a/day ($22,000/2,400 gal/day capacity) capital  cost esti-
mate used here.   Operating costs for the larger system were  also substantially
lower, falling in the range of $7.4-9/1,000 gallons.   The higher cost esti-
mates for both capital and operating were used throughout the remainder of
this analysis.
       Installation costs for this equipment have been estimated  at $6,000,'
not including a storage facility.  The equipment selected is expected to
occupy 80 square feet of deck space.
                                                        36
 Model  Plant
  Capital Cost
of Water Treatment
     Operating
        Cost
Installation
    Cost
                   Total
 Offshore
     4,000 hp    $21,000
            $/kW
            $7.00
$/1000 gal   $/yr
 $11.00    $9,640     $6,000.00
 7.2.1.1.5  Electric generating station - The model plant selected for analysis
 in  this  application  is a 66 MW turbine installed at an electric generating
 station.   Operating  tiire is assumed to be 500 to 3,000 hours per year.
 Water treatment  system selection was based on a system currently being
                                    39
 installed  on  a turbine in  New Mexico   which was designed to meet a 0.3
 lb/10 Btu NO limitation, approximately equivalent to 75 ppm on liquid
 fuel.  The system selected by this utility is a R.O.  unit preceded by a
 pretreatment  system  designed to control the pH.  Water used has a TDS of
 467 ppm  and is not high  in other constituents.  The R.O. unit is followed
                                     7-18

    -------
by one mixed bed de-ionizer which feeds to a storage tank.   Total  bid
cost for this system delivered.to New Mexico was $74,000 adding the $26,000
for the required pumps and storage facilities, system cost 1s equivalent
to $1.50/kW of capacity.   The capacity of this system is 60 gpm, somewhat
in excess of the 42 gpm requirement estimated by the turbine manufacturer.
       Another system on three turbines equalling 192 MW was installed in
                                                                   oo
the spring of 1975 at a total capital cost of $210,000, or 1.09/kW.    This
system uses a two bed de-ionizer (DI) system in place of a R.O. and mixed
unit.   A storage tank of 50,000 gallons is provided to serve during regenera-
tion of the DI units.  The system is designed to meet an emission  rate of
75'ppm NO .
         ^
       San Diego Gas and Electric operates a different type system.  Due to
the high TDS level of San Diego water, a prefiltering system was required.
San Diego also chose to use a dual bed system followed by mixed bed polishers.
The system shown, designed for 25,000 gallons per day, or 17 gpm,  serves
20 MW of turbines; any excess is used as boiler makeup water.  The capital
                                        39
cost of this system was $70,000 in 1971;   current operating costs are
                                                         40
$2.90/1,000 gallons, excluding amortization of equipment.    This  operating
cost has been assumed for the model plant, although it is felt to  be somewhat
high.   Other costs reported for water systems similar to the one proposed
are $1.25/1,000 gal.
       Other capital and operating cost have been estimated for expected
                                                                 41
turbine installations in different utilities.  One survey by ASME    shows
capital costs ranging from less than $1.00 to $10/kW with an average of
$4.86/kW.  What factors and system types were considered are not clear.
                                    7-19

    -------
Also unclear are the type of system proposed or the type of input water used.
Operating costs for the treatment system ranged from 0.006 mills/kWh to
0.567 mills/kWh, with one at 10 mllls/kWh.   These operating costs work out
to be $.15/1,000 gallons to $14.85/1,000 gallons.  Again, the reason for
the difference is not clear.
Model P^nt

Capital Cost
of Water Treatment
Total $/kW
Operating
Cost
$/100Q__gal $/yr
Installation
Cost

Electric Utility
  66 MW            $100,000     $1.50     $2.90     $7,300
$70,000.00
7.2.1.1.6 Other costs and variables - In addition to the actual costs of
the water purification equipment, additional costs are incurred due to the
modification of the turbine caused by installation of water injection
nozzles and other equipment.  These costs, estimated by the manufacturers,
vary widely.  Table 7.5 shows the costs reported by various manufacturers.
The reasons for the difference in reported costs may be due to assumptions
that were not specified by ttie manufacturers.  For this analysis, the costs
reported by Solar are used for the smaller turbines and the costs reported
by Westinghouse in its proposal to New Mexico Electric Company are used for
the larger turbine.  The choice of Westinghouse over G.E. was made for two
reasons:  first, the cost for the Westinghouse turbine represents an actual
bid cost to modify a turbine for water injection and, second, since it was
not clear how much of the equipment included in the water quality treatment
system  (pumps, piping, etc.) may have been included as turbine modifications
by other manufacturers, it was felt best to use the entire cost from one
project where possible.
                                    7-20

    -------
          TABLE 7.5
COST OF TURBINE MODIFICATIONS
Manufacturer
Solar
G.M.
G.E.
Unknown
Westing house
Ref.
42
43
t
44
45
46
47
48
Size
noo HP
3800 hp
3800 hp
66 MW
22 MW
22 MM
65 MW
66 MW
Cost
i
2,900
3,900
10,000
70,000
70,000
20,000
53,000
i
45,400
$/hp or
2.64
1.03
2.63
1.06
3.18
.91
.82

               7-21

    -------
   i    Turbines located in cold weather climates may require further equip-
ment to prevent freezing of the water.  This requirement, however, is not
at all certain.  All turbines currently in use require heated buildings.
If water storage is not required, thermal heating within the building should
be sufficient to protect against freezing.  If storage is considered a
necessity, then the exhaust gas from the turbines could be used; only ducting
and venting would add to the cost.  Total costs for these precautions are
expected to be insignificant if the turbine is used in a baseload application.
If the turbine were to be used as an emergency generator or peaking unit to
supply electricity, it is assumed that electric heat would be available at
minimal additional  cost.
       The quality of the incoming water could also affect some of the costs
reported here.  Both capital and operating costs could be significantly
impacted if either very poor (both TDS and other constituents high) or better
(low TDS) water were available.  Estimated changes in capital costs ar? in
the range of + 15 percent.
  i
       A significant variable in assessing capital and operating costs is
variation in water injection requirements between manufacturers.  Tables 7.6
and 7.7 show that variation.
       One other capital cost which could be attributed to an NO  emission
                                                                A
standard, if based on mass of pollutant/unit of power output, is a requirement
to measure to power output of each turbine.  Based on information supplied by
                 49
one  lanufacturer,   the installed cost of this equipment on a large turbine
is between $13,000 and $17,000.  Little,  if any, operating and maintenance
penalty will be incurred with this equipment.  No data have been found con-
cerning the application of this equipment on turbines less than 20,000 hp.
                                   . 7-22

    -------
                       TABLE 7.6
       'VARIATION IN WATER FLOW REQUIREMENTS FOR

                    75 Dom EMISSION LEVEL
Manufacturer
Solar

G.M.
Turbodyne
G.E.

Westing house
TPM
Size
• 1100 hp
3830 hp
3900 hp
65 MW
66 MW
20 MW
66 MM
£6 MW
Flow
(gpm)
0.25
1.66
0.85

22
5.8
42
100 gpm
Model
Plant
2
2

3

3
3
'* Model Plant:  1 - Standby  Power
                2 - Baseload  application  (industrial  or pipeline)
                3 - Utility  power  plant  ,
                         7-23

    -------
                                  TABLE 7.7
                   VARIATION IN WATER FLOW REQUIREMENTS
                         FOR 125 ppm EMISSION LEVEL
Manufacturer

Solar


G.M.

Turbodyne

G.E.


Westinghouse
      Size
Flow
Model Plant*
1100 hp
3830 hp
3900 hp
65 MW
66 MW
20 MW
0
0
0
50
8
4.2
1
2
2j
3
3
3
     66 MW
 25
      Model Plant 1
                  2
                  3
Standby
Baseload (industrial or pipeline)
Utility powerplant
                                     7-24

    -------
7.2.1.1.7 Change in control costs for a 100 or 125 ppm NO  emission level -
                                                         A
Control costs for either the 100 or 125 ppm emission level are expected to
be quite similar.  No additional control will probably be required for the
standby or industrial turbines.  The flow rate reduction required for the
utility application is not expected to significantly affect the overall
control cost for those manufacturers who would require wet type controls
to achieve either the 100 or 125 ppm emission limit.
7.2.1.1.8 Summary of water injection control costs - Table 7.8 shows a
summary of the cost data selected for the model plants.
7.2.1.2 Steam Injection - As also discussed in Chapter 4, the injection of
steam into the combustion chamber can successfully control emissions of NO .
                                                      s                   *»
Injection of steam (on a mass comparison with water) may not be as efficient
as water injection in reducing NO  emissions; up to three times as much steam
                                 rt
might be required, according to one turbine manufacturer.    On the other
hand, other manufacturers report equal effectiveness for water and steam.
Steam injection results in about a 2 percent improvement over water injection
                                                                         51
in the heat rate of the turbine and a comparable increase in power output
The drawback to steam injection is that, unless a ready supply of steam is
available, the cost for raising steam to the proper temperature and pressure
is considerably higher than the costs for providing water for injection.
                                      52
According to one turbine manufacturer,   this cost could be three times higher
than the cost of water injection, assuming the installation of an Intermediate
pressure boiler specifically for the injection steam.>
       In certain circumstances the injection of steam seems to have potential
applications.  If the turbine is part of a combined cycle operation, it is
possible to take some of the steam-from the boiler portion of the combined
cycle plant for use as the steam input.  Of course, this would detract from
the overall efficiency of the combined cycle, but in many instances the overall
                                  • 7-25

    -------
                                  0)
                                  o
                                  ret
                                  0.
                 O
                 CM
                 O
                 00
                                   s
                                              8
        o
        o
                  o
                  o
                   *
                  CVJ
                  o
                  o
                  o
                   *
                  vo
                                   o
                                   o
                                   o
 o

    4->
 |3 
                                                                                   en
                                                                                   ID

                                                                                   s
                                                                                   v>
     QCO
CO   t/0»-i
     i/iuu

          $-


 «X>    4-> W
 O     V*
O  U  >>
    
 C  (0 4-> i—
•r- IZ  C  
                                              o
                                              o

                                              en
o
o
  •
o
CM
                                                       CM
                                                       CM
o
o
  •
o
CM
                  o
                  o
                  CM
                  in
                  o
                  o
                   •
                  CO
                  o
                  o

                  vo
                  CO
                  VO
                  CM
                                               o
                                               o
                                    o
                                    o
                                     •
                                    o
                                    o
                                    co
                                                    o
                                                    CD
                                                      •

                                                    CM
                                                                                  •o
                                                               (O

                                                              I
                                                                                                             in
                                                                                                             o
                                                                                                             o

                                                                                                             o
                                                                                                             o
                                                                                                             o
                                                                                                              A
                                                                                                             00
     _
     UJIX
     QO
                                                              •o
                                                              •o
                                is?
                           O  I-  O
                          o  CJ  re *»*>
                              4-> U.
                                                       §
                  o
                  o
                  o
                  o
                                    o
                                    m
«*•       F-
                                                              CM
                                                                                                             >
                                                               O
                                                               ex
                                                               U)

                                                               (O

                                                               •M
                                                      0)
                                                      4->
                                                      
    -------
economics may justify the use of the steam.   Steam produced by other onsite
boilers either at an electric utility or at other installations may also be
used for steam injection.  For example, one turbine user has recommended the
use of steam injection solely to enhance output.
       In summary, steam injection is an effective alternative to water
injection which may be used in circumstances where the particular site
economics would justify the choice.   None of the model plants have been
assumed to use steam injection, since it is not felt that steam injection
would be used unless particular site economics justified the choice.
7.2.1.3 Catalytic Exhaust Gas Treatment Control - One vendor has developed
a catalytic exhaust gas cleanup control device to remove NO  from turbine
                                                           ^
         53
exhausts.    As in the case of steam injection, this device appears to have
use in certain applications, especially where a fuel high in fuel-bound
nitrogen was used or where water was not available.
       Estimated costs for this system applied to a 30 MW turbine are in the
vicinity of $135,000 ($4.50/kw) which is higher than water injection costs
for a similar sized machine (i.e., approximately $1.5-$2.0/kw).  In addition,
the system requires the use of considerable amoun-s of ammonia and a method
of storing and handling it.  Also, operation of the system causes about a 2
                                   54
percent decrease in turbine output.
       The catalytic exhaust cleanup control system is probably suited for
certain applications.  Its costs and operating characteristics are not pre-
sently more favorable than the costs of a water injection system; for purposes
                                                     I
of economic analysis, it will not be considered as a separate option, since it
is not expected to be used unless the cost is less than the cost of a water
injection system.
                                    7-27

    -------
7.2.1.4 Dry Controls - By modifying the design of the combustor or by re-



vising the methods of firing, lesser amounts of NO  can be formed.  This
                                                  A


method of control is called "dry" control  technology, since 1t does not re-



quire the addition of water or steam to meet NO  limitations.   To varying
                                               A


degrees most manufacturers have incorporated dry controls and  have reduced



gas turbine emissions relative to uncontrolled levels.  Some machines (see



Table 7.2) have reached the level of the most stringent standard and machines



from at least one manufacturer in each size range can meet the 125 ppm level



using the current level of dry control.  Some of the smaller machines on the



other hand, do not incorporate dry controls, but they may meet emission levels



due to low firing temperatures and compression ratios.

  \

       Many gas turbine manufacturers felt that if current R&D programs were



successful, future improved dry control sufficient to meet the most stringent



proposed standard should cost about the same as current levels of water in-



jection.  Therefore, it was assumed that no additional dry control was needed



to comply with the proposed standard levels.  Rather, as in the case of steam



injection, water injection was assumed for the model plants because it was



felt that any added degree of dry control  must cost less than  current water



injection.



7.2.1.5 Effect of Fuel Specifications in NO  Control -- The predominant fuel
                                           /\


currently used by gas turbines is either distillate fuel oil or natural gas.



In fact, between January and April 1975, no utility turbine in the U. S. burned



any fuel except distillate or natural gas.  As discussed in Chapter 4, these



trends will probably continue in the future.



       Both natural gas and distillate fuel oil are relatively clean fuels



which are low in both fuel nitrogen and sulfur.  Sulfur will be discussed in



the next section.  Fuel bound Np is capable of significantly impacting the
                                    7-28

    -------
ability and cost of turbine compliance with NO  limitations.  For example, a
                                              x          ;!
fuel containing 0.1 percent N2 (very high for distillate and average for
residual) would add about 32 ppm of NOX to a turbine burning 100 gpm of fuel;
a fuel containing 0.046 percent N£ (higher than almost all distillates) would
                                       55
add about 16-17 ppm to the stack gases.    Wet controls can compensate for
these levels, although the water or steam injection rate would have to in-
crease,   There should be no.problems with catalytic exhaust cleanup control.
If relatively clean fuels are used, fuel bound Np should not be a problem.
       If turbines were to burn residual fuel oil, a cleaning facility to
remove sodium, ash and other water soluble impurities would have to be
installed to treat the oil prior to its use in the turbine.  Based on dis-
cussions with two manufacturers of this equipment, the cost to treat residual
fuel for a turbine using 100 gpm (about 60-70 MW) was between $400,000-
750,000 plus installation, estimated at a minimum of 50 percent of the capital
V
cost.  Operating costs would not be expected to be high ($2/1,000 gallons)
except where large amount of vanadium were present.  One manufacturer quoted
a price of $20-25,000/day for a very large turbine installation burning a
high vanadium residual oil.  These costs do not include any additional costs
which might be required to provide facilities for handling residual fuel oil.
These facilities might include heavier oil pumping systems, turbine modifi-
cations, oi'i ncdcers, etc. which may already be present if the turbine is
located at an oil-fired power plant.  These costs were confirmed by several
gas  turbine users who reported installed costs for fuel treatment systems
ranging from  $1,300,000 to $1,700,000.56'57/ Based on the above cost data
both manufacturers of oil treatment facilities indicated that they did not
                                    7-29

    -------
foresee a large amount of residual  fuel  being used  to  fire  turbines  in  the
United States.  To confirm this assessment, a simplified discounted
cash flow analysis war performed using the following equation:
              n    (l-T)(Rt - Ct - DJ + Dt - Kt
       PV =  v-^  - S - L_t - i - L.
             2-,         (1 + r)*
              t=l
 Definition of Terms
       K = capital cost and depreciable installation costs
       R = gross revenues
       T = applicable tax rate
       C = operating costs, overhead charges, working capital requirements
       D = average or annual rate of depreciation
       t = year of cash inflow or outflow
        n = useful  life of investment
        r = discount  rate  (i  =  internal  rate of return  when  PV =  0).
  By setting the present value  equal  to  zero, this  equation  can be solved for
  the price differential  between residual  and distillate fuel  oil  required
  to justify the investment.  The formula  was solved using the following assump-
  tion and for two levels  of operation (i.e., 500 and 2000 hours  per year):
        K = $1,000,000
        C = $2/1000 gallons:  $4704 9 500 hours and  $18«16 
    -------
At 500 hours of operation per year, the required price differential would be



roughly $5.22/barrel or $.126/gallon.  At 2000 hours per year the differential



decreases to $1.38/barrel or $.033/gallon.  Historical price differentials be-


                                                                             co

tween residual oil and distillate have been about 2-3(/gal1on at the refinery



and currently are about 2
    -------
7.2.2 Cost of Control  ofjiulfur Oxides for New Turbines



       The only source of sulfur oxides from a gas  turbine  is  sulfur  contained



in the fuel.   As discussed in Section 7.2.1, few if any  turbines  the  U.  S.



are expected to burn residual.   In this section, the costs  attributable  to



burning 0.28% sulfur oil  (about 0.3 lbs/106 Btu) versus  0.74%  sulfur  oil



(about 0.8 lbs/106 Btu) are analyzed.  As will be discussed,  the  0.28% sulfur



level is typical of most distillate fuel  oils, while the 0.74% level  represents



the New oource Performance Standard for oil-fired steam  generators.



7.2.2.1 Residual Fuel  Oil -- Costs for residual oil desulfurization  are



dependent on the type of crude used to produce the  residual  fuel  oil.



Tables 7.9 and 7.10 show the costs for various crude types  and desired sulfur



levels.  Based on a previous study of the electric  utility  industry,    a dis-



tribution of imported product by source and demand  by sulfur content were



developed.  These are shown in tables 7.11 and 7.12.  The demands shown are



based on current State regulations and Federal new source performance standards.



The average cost to reach these levels was $0.99/bbl or 2.4
    -------
 * Reference 60
                              TABLE  7.9
                    DIRECT DESULFURIZATION COSTS
                      .(1974 DOLLARS  PER BARREL)*
Desired
Sulfur
 Level
           Crude Source
Arabian Light            Texas Sweet
 0.2%
 0.3%
 0.5%
 0.7%
 1.0%
 1.5%
 2.0%
 2.5%
 3.0%
  1.73-2.00
  1.60
 '1.46
  1.33-1.77
  1.18
  0.90
  0.61
  0.35
  0.00
1.23
1.13
0.96
0.74
0.49
0.00
0.00
0.00
0.00
                            .   7-33

    -------
                                    TABLE 7.10
                           INDIRECT DESULFURIZATION COSTS*
                              (1974 DOLLARS PER BARREL)
Desired
Sulfur
Content
0.2%
0.3%
0.5%
0.7%
1.0%
1.5%
2.0%
2.5%
3.0%
Oil Source! Venezuela
Process









Visbreak** Coking Heavy Oil
+ Vacuum HDS Crack
$1.36 $1.64 $1.66
1.28 1,54 1.63
1.15 1.36 1.50
0.95 1.18 1.36
0.81 0.92 1.14
0.47 0.52 0.59
0.13 0.15 0.17
0.00 0.00 0.00
0.00 0.00 0.00
California 1 .,
• 	 L Heavy
Visbreak** Coking Oil
+ Vacuum HDS Crack
$1.48 $1.73 $1.79
1.34 1.64 1.77
1.33 1.54 1.73
1.15 1.41 1.64
1.08 1.24 1.43
0.79 0.96 1.00
0.55 0.67 0.69
0.31 .0.37 0.38
0.00 0.00 0.00
 *  Reference 60.
** Costs reflect discard  of  high-sulfur pitch.
                                         7-34

    -------
                                       TABLE 7.11
                           MIX OF CRUDE OILS USED TO PRODUCE
                             RESIDUAL OILS IN U.S. MARKETS
                             (MILLIONS OF BARRELS PER DAY)*
   From Imported Product
     Venezuela
     Arabia (light)

   From Domestically Produced Crude
                                   i
     Texas  (sweet)
     California (sour)

   From Imported Crude
2.13
     Canadian/Algerian
     Arabia  (light)
         Total  Demand
                      ***
 ,87
 .70
                           (1.92)**
                           ( .21)
                           ( -78)
                           ( .09)
                           ( .
                           ( .52)
3.70
  *   Reference 60.
 **   Numbers in parenthesis refer to  the  ambient cf oil (millions of barrels per day)
***   Approximately equivalent to Texas  sweet.
                                            7-35

    -------
                                   TABLE 7.12

                      DISTRIBUTION OF RESIDUAL FUEL OIL*
                                 DEMAND IN 1980
Sulfur
Content %
0
0.2
0.4
0.6
0.8
1.2
1.8
2.2
'
- 0.?
- 0.4
- 0.6
- 0.8
- 1.2
-' 1.8
- 2.2
- 2.8
>2.8
UUIICLIIU
(IP6 BBL)
1.64
97.43
217.9
114.5
196.5
3.2
28.9
13.3
63.8
Percentage
0.2
13.2
29.6
15.5
26.6
0.4
3.9
1.8
8.8
: Cumulative
Percentage
0.2
13.4
43
58.5
85.1
85.5
89.4
' 91.2
100
* Reference 60.
                                      7-3fi

    -------
State requirements are expected to be much smaller than this.   Current costs
for low sulfur (0.3 %) oil are about $1.30/bbl  over high sulfur o1l.61^
7.2.2.2 Distillate Fuel 011 — The majority of  distillate fuel  oils  sold 1n
the U.S.  will  meet both a 0.3 and 0.74% sulfur  regulation with  no further
treatment.   Some fraction of the distillate, however,  may have  to be desulfur-
ized prior to  use if a 0.3 standard were imposed.   Table7.13 shows the distri-
bution by sulfur content of distillate fuel delivered  to power  plants in iy74.
       Based on the above data, it '.s clear that the economic impact of a
0.74% limitation is effectively zero, while some desulfurization could be
required if a  0.3 standard were set.
       As the  potential demand for low-sulfur distillate oil  by turbines is
   \
a relatively small proportion of total demand for middle distillates, and
since a large  proportion of this oil  is below 0.3% sulfur, the  demand can
probably be met by blending distillates or by selectively purchasing low-
sulfur crude.   The maximum premium for this fuel can be estimted as  the cost
to desulfurize a higher sulfur distillate down  to a 0.3 level.   Little data
exist on desulfurizing distillate fuel, but the cost can be approximated
closely by the cost to desulfurize gasoline.
       Desulfurization of gasoline can be used as a proxy for desulfurization
of distillate  because the higher (>0.3%S) sulfur distillate is produced as a
stream from the catalytic cracker and gasoline is desulfurized  by treating the
feed to the cat cracker.  The estimated cost for this is about  H/gal.
       If, as  has been suggested, gasoline is desulfurized to alleviate sul-
                                                      i
furic acid from automobiles, the cost attributable from low-sulfur distillate
would be reduced to zero.
       Overall, the maximum differential impact between a 0.3 and 0.75% sulfur
limitation on  distillate fuel oil is expected to be less than 0.5tf gal, or less
than a l%*increase in fuel costs for turbines.
                                    7-37

    -------
*  Reference 60.
                                     TABLE  7.13
                           DISTRIBUTION  OF  DISTILLATE OIL
                              BY SULFUR  CONTENT  -  1974*
 Sulfur Coi tent
      (*)
Percentage of
    Total
Cumulative
   0.1  - 0.2
   0.2  - 0.3
   0.3  - 0.5
    >0.5
     1.9
    45.4
    12.6
    25.4
    14.2
    1.9
   47.3
   59.9
   85.3
  100.00
                                           7-3R

    -------
7.2.2.3 Flue Gas Desulfurization (FGD) -- In theory, 1t should be possible
to apply FGD systems to turbine in a fashion similar to power plants.  In
practice, the gas flow rate of a turbine relative to a steam generating
power plant is extremely high.  The estimated cost of a FGD system can be
estimated on the basis of the gas flow rate to the system.   The flow rate of
the model electric utility turbine is about 1.5 x 10  ACFM or equivalent
to a 500 MW steam generator.    Installed costs for this size are
estimated at  between 26 and  30 million dollars or 2 to  3 times the cost of
            64/
the turbine.     Needless to  say, it  is  not expected that many turbine
users would decide to use FGD systems.
7.2.3 Cost of Particulate Control for New Turbines
       As discussed in Chapter 4, a wet  scrubbing system has been developed
to reduce particulate emissions for gas  turbines.  Based on data from the
manufacturer   the installed  cost of this system has been estimated for
the model plants.  Table 7.14 shows these results.  It  is obvious that for
the smaller turbines the cost of this equipment is prohibitive.  For the
utility application the capital cost is  quite high but  other factors must
also be considered.   Makeup water use by the system is  about 500 gpm; the
space requirement is very large for the  cooling tower alone and, coupled
with the space needed for the scrubber itself, can amount to over twice the
space needed  for the turbine  itself.
7.2.4 Cost of Control for Modified Turbines
       As discussed in Chapter 5, the probability is quite high that few,
if any, turbines will fall within the definition of modifications.  However,
if any turbines did fall within the modification definition, the cost of con-
trol would not be different  than for new turbines, since the area for control
(either wet or dry control)  is within the combustor can which is commonly
changed daring any type of change which  might be considered a modification.
                                    7-39

    -------
                               CM
                               CM
                                        ro
                                                 in
     to
     H-

     
                                                                     0)
                                                                     (/>
                                                                     C
                                                                     0)
                                                                     Q.
                                                                     X
                                                                          C    u_  J-


              "    I
                 "
                     u
                          o
                          o-
                          o
                           A
                          o
                          CM
                          CM
o
o
CD
  ft

O
CTl
CM
O
o
o
  •k
O
r-»
CD
 O  «*-
                     O
                    •»—

                    4-*

                     10
                     O
                    a.
                    a.
                    
                                  o  *o
                                  O   to
                                  o


                                  sT  j?
                                  Q)   C
                                  JQ   O
                                  JQ
                                                                     O  »
                                                                     (/>  O
                                  
    -------
7.3 COST ANALYSIS OF GAS TURBINE EMISSION CONTROL SYSTEMS
                                                    •
7.3.1 Introduction
      The cost estimates presented below cover four major gas turbine
applications:
      (1)  standby emergency power generation units;
      (2)  industrial uses such as internal power generation
           or pipeline compressor stations;
      (3)  off-shore drilling platform pumping and compressor
           units; and
      (4)  utility peak load generating stations.
The purpose of this section is to analyze the potential impact of the pro-
posed standards of performance on the cost of generating power in these
applications.  To the extent possible, the data used in estimating incre-
mental control costs were taken either from aata provided by users of gas
turbines currently using water injection to control NO  emissions or from
                                                      /\
estimates provided by manufacturers of gas turbines and manufacturers of
water purification equipment.  These data were subjected to detailed analyses
and are considered to be reliable.  However, since these data were
estimates, the parameter values were widely varied to determine,
the sensitivity of the resulting aggregate cost estimates to changes in
the value of specific subcomponents of the emission control system.
      A mathematical model was constructed to assist in estimating the cost
(mills/kilowatt hours) and percentage impacts of the proposed standards of
performance.  The model was also used to test the sensitivity of the esti-
mates to changes in some of the underlying assumptions.  Contained within the
model were a number of parameters which impacted directly on the unit costs
                                     7-41

    -------
(mills/kilowatt hours) of operating gas turbines.^  Some of these para-
meters may be considered to be independent of the application;  others
depend on the location, size and use to which the turbine will  be put.,
7.3.2 Baseline Cost Estimates
      Gr.s turbines are widely used as standby generating stations by
hospitals, telephone companies and other institutions where power inter-
ruptions cannot be tolerated, by industry for internal power generation,
by pipelines at compressor stations and by the utility industry to meet
peak load requirements.  In all cases, the cost of operating the turbine is
a function of the application.  Baseline costs for each of the model
applications are provided below.  The costs have been estimated in terms
of fixed capital charges, fuel costs and operating and maintenance expendi-
tures.
       Although some  states and localities have regulations requiring NO
                                                                        A
control on gas turbines, baseline costs were estimated assuming no exist-
ing control requirements, because few specific regions have control
requirements for gas  turbines.  In addition, since the assumption tends
to overstate the incremental cost, it can be viewed as conservative.
7.3.2.1 Parameter Values
7.3.2.1.1 Fixed capital charges -- The fixed capital  charge is  the percentage
of the total capital  cost including all depreciable installation charges
which must be recovered each year over the life of the investment to pro-
vide an acceptable return on invested capital.  The aqtual percentage which
muse be recovered will depend on the type of cost accounting system used,
the method used to depreciate capital stock for tax purposes and whether
installation and other relatively fixed costs are included with the purchase
                                      7-42

    -------
price of the equipment.  Typical fixed capital charges used by Industry
vary from 15 to 25 percent for plant and equipment expenditures with
useful lives in excess of ten years.
      In the cost model, three fixed capital charge rates were used.
Twenty percent was used as an average value in most instances.  To test
the sensitivity of the result, the charge rate was varied from 15 to 25
percent.  It is doubtful that a fixed charge rate higher or lower than
these values would be appropriate given the cost of funds in the existing
capital market and the useful life of the equipment.
7.3.2.1.2 Fuel costs — Depending en the application, fuel costs may re-
present the most important element in the overall cost structure for gas
turbines.  Clearly, this will not be the case for emergency standby units
which are seldom, if ever, operated more than 200 hours per year.  Fuel
costs  in this application  probably never account for more than 20 percent
of the annual total cost of the system.  In almost all other applications,
with the possible exception of utility peaking plants operated less than
1,000 hours per year, fuel costs will account for over 50 percent of annual
costs.
      Two fuel costs were used in this analysis.  Distillate fuel oil, which
accounted for roughly 60 percent of the energy consumed by the electric
utility industry in turbines in 1974, was priced at $2.18 per million BTU.
This price, based on the current selling price of distillate fuel oil  in
the Gulf Coast region of the country,66 is roughly equivalent to $12.70
per barrel, or $.30 per gallon.  The other fuel cost was the average
city-gate price for natural gas which, during the first quarter of 1975,
was equivalent to $.66/MMBTU.67/
                                   .7-43

    -------
       With regard to the cost evaluation, in most cases the distillate
fuel oil price was applied for two reasons.   First, natural  gas is 1n
short supply.   Industrial customers, on firm as well as interruptible
contracts, are being .everely curtailed.   With the exception of electric
utilities located in the intrastate market,  the outlook for future natural
gas supplies in the utility industries is even less favorable.   It is un-
likely, therefore, that new turbine units will be brought online using
natural .jas as their primary fuel.  More likely, new turbines that have
gas-firing capabilities will be equipped to fire oil in the event of
future curtailment.  Secondly, in intrastate markets where natural gas is
still available, the price is approaching or even exceeding the price of
oil.  Table 7.15 shows a summary of intrastate natural  gas contracts signed
during the period January 1, 1974 to January 9, 1975.  While in the quoted
new contract prices vary considerably, they are considerably higher than the
controlled interstate gas prices and likely to continue to increase.  In a
free market, natural gas prices can be expected to increase to levels com-
parable to substitute fuels.  The only major fuel which is clearly substi-
tutable for natural gas in gas turbines is distillate fuel oil.
7.3.2.1.3 Operating and maintenance charges — Gas turbines have relatively
low operating and maintenance requirements compared to other equipment.
This is one of their primary advantages.  Data from the Federal Power Com-
mission indicate that for utility systems operating and maintenance costs,
exclusive of fuel costs, are generally less than one percent of the installed
                            fn
capital cost of the turbine.    Data on a randomly selected group of utility
turbine installations are presented in Table 7.16.  In seven of the nine
                                    7-44

    -------
                                TABLE 7.15

                          NATURAL GAS INTRASTATE

                                GAS PRICES

                       (Jan. 1, 1974 - Jan. 9, 1975)


                               .  PRICE i/MCF
        FPC Area                      High        Average        Volume Mcf
                                           \
Appalachian-Illinois ]_/               75.00         52.95          7,164,886
Other Southwest 2J                   152.53        116.93         24,251,836
Southern Louisiana 3/                121.19         87.99         25,031,297
Texas Gulf Coast 4_/          .        207.22         83.09        256,864,228
Permian Basin 5/                     146.79         89.63        185,258,708
Hugoton-Anadarko 6_/                  110.60         73.99         56,770,320
Rocky Mountain 1J                     77.00         27.43          3,108,302
Other Areas 8/                       243.89        131.76          3,941,964
Source:  Statement of John Nassikas, FPC, before the House Committee on
	Interstate and Foreign Commerce, July 14, 1975.

\J  Areas of New York, Pennsylvania, Ohio, West Virginia, Maryland,
    Virginia, Kentucky, Illinois, and Indiana.
                                                                    •j
2/  Areas of Mississippi, Louisiana, and Texas.

y  Southern Louisiana.

4/  Areas of Texas.

5/  Areas of Texas and New Mexico.

6/  Areas of Texas, Oklahoma, and Kansas.

TJ  Areas of Louisiana, Utah, New Mexico, Colorado, Wyoming, Nebraska,
    Montana, North Dakota, and South Dakota.

8/  The remainder of the Continental United States not included in any
    of the other producing areas.
                                 .  7-45

    -------
                              .   TABLE  7.16
                     UTILITY OPERATION AND MAINTENANCE
                        EXPERIENCE WITH GAS TURBINES
Plant Name
Unit Size

  (MW)
Installed Cost

    ($/KW)
     Operating and
Maintenance Expenditure
                                                % of
                                     Installed
                                     Cost
                                   Hills/
                                    KwH
Simple Cycle
Yuma
Intercission City
Turner
Fox Lake
Sycamore
Meraniec
Combined Cycle
Sterlington
Croydon
Gilbert
192.0
340.0
181.0
2Q fi
t- J • \J
157.5
58.9

523.4
546.0
205.2
                     85
                     73
                     86
                   $ 68
                   $ 89
                   $ 91
                                   $
                                   $
                                   $102
                                   $115
                                   $160
                    1.
                    0.
                    0.
                    0.
                                                   0.2
                                                   0.3
                                   2.7
                                    .2
                                    .7
                 3.3
                 0.7
                 1.3
                 2.2
                 1.0
                                      .8
                                      .3
                                      .4
Source:  Form 12,  FPC data for 1974.
                                     7-46
                                                                                    r*r

    -------
cases, operating and maintenance costs were less than 1 percent of the
installed capital costs when Utility systems were considered.
            For industrial  applications, a higher average operating and main-
tenance cost was used, because these units are generally operated continuously
and require a major overhaul every five or six years? '  The cost of the over-
haul ranges from 15 to 25 percent of the installed cost of the turbine.  For
purposes of cost evaluation, a 5 percent operating and maintenance cost per
year was assumed.  Four percent is attributable to the required overhaul and
1 percent for general operating and maintenance expenses.
            Units in standby service generally receive little maintenance.
In the absence of a power failure, they may be activated once or twice a year
to ensure that they will operate properly in the event they are needed.
since data on actual operating pxpcriece were lacking, operating and maintenance cc
were assumed to be approximately 1 percent of the installed purchase price.
7.3.2.2  Cost Estimates -- The parameters discussed above were used to establish
baseline costs for the model plant applications.  Table 7.17  shows the es-
timated costs on a mills per kilowatt hour basis.  Heat rates (BTU/kwh) and
installed costs per kilowatt were based on manufacturers' specifications and
data reported to the
         The cost estimates presented in Table 7/17 vary within wide margins.
Standby applications have the greatest variability, with costs ranging from
265 to 1189 mills per kilowatt hour.   Capital costs' dominate, accounting for
over 80 percent of the cost per unit of output in all  cases.   Fuel  costs are,
for the most part, quite small compared to other gas turbine  applications.
                                   7-47

    -------
                                 TABLE 7.17


                          BASELINE  COST ESTIMATES1
                        FOR MODEL PLANT APPLICATIONS
A.   Standby Units
1.   Operating Parameters
          Unit Size
          Installed Capital  Cost
          Heat Rate
          Fixed Capital Charge
          Operating and Maintenance
          Fuel Costs
                                     350 hp, 1100 hp
                                     $2207Kw
                                     15,500 BTU/Kwh
                                     20% of Installed Cost
                                      1% of Installed Cost
                                     $2.18/MMBTU (#2 oil)
    Estimated Baseline Costs
Mills/Kwh

Allocation of Cost
   (% Total)

(1) Fixed Capital Charges

(2) 0 + M

(3) Fuel Cost	
B.
     Total

Industrial  Applications
1.  Operating Parameters
                                Operating Hours -- per Year

                                40     80     120     160     200

                              1188.8  611.3  418.8   322.5   264.8
92.
4.
2.
100.
4
6
8
0
90
4
5
100
.0
.5
.5
.0
87.
4.
8.
100.
6
4
0
0
85.
4.
10.
100.
2
3
5
0
83
4
12
100
.0
.2
.8
.0
          Unit Size
          Installed Capital Cost
          Heat Rate
          Fixed Capital Charge
          Operating and Maintenance
          Fuel Costs
                                      3000  Kw
                                      $120/Kw
                                      13,200 BTU/Kwh
                                      20% of Installed  Cost
                                      5%  cf Installed Cost
                                      $2.18/MMBTU  (#2 oil)
1
 All cos1>s are estimated on a before tax basis.
                                    7-48

    -------
                                TABLE 7.17 CONTINUED
2.  Estimated Baseline Costs
                              Hours of Operation  per Year
2000 3000
Baseline Costs 43.8 38.8
(Mills/Kwh)
Allocation of Cost
(% Total)
(1) Fixed Capital 27.4 20.6
Charges
(2) 0 & M 6.9 5.2
(3) Fuel Costs 65.7 71-2
Total 100.0% 100.0%
C. Utility Application
1. Operating Parameters
Unit Size
Installed Capital Cost
Heat Rate
Fixed Capital Charge
Operating and Maintenance
Fuel Costs
2. Estimated Baseline Cost

200 500
Baseline Costs
(Mills/Kwh) 180.0 85.5
Allocation of Cost
(% Total)
(1) Fixed Capital 83.3 70.2
Charges
(2) 0 & M 4.2 3.5
(3) Fuel Cost 12.5 26.3
Total 100.0% 100.0%
4000 6000 8000
36.3 33.8 32.5

16.5 11.8 9.2
4.2 3.0 2.3
79.3 85.2 88.5
100.0% 100.0% 100.0%


66,000 Kw
$150/Kw
10,300 BTU/Kwh
20% of Installed Cost
1% of Installed Cost
S2.18/MMBTU (#2 oil)

Hours of Operation per Year
1000 2000 8000
54.0 38.2 26.4
-
55.6 39.3 14.2
2.8 2.0 .7
41.6 58.7 85.1
100. 0% 100.0% 100.0%
                                   7-4Q

    -------
       Baseline costs for the industrial  applications were the least sensitive
to changes in operating hours, since fuel  costs dominated and, to a signifi-
cant degree, the heat rate was independent of the number of hours of opera-
tion.  Baseline costs ranged from 43.8 mills per kilowatt hour for a turbine
used in a single shift industrial operation (2,000 hrs/yr) to 32.5 mills per  '
kilowatt hour for a unit operated continuously eleven months (8,000 hrs) of
the year.
       Simple cycle turbines are used by utilities and generally confined to
peaking operations.  Few of these units are operated more than 2,000 hours
per year.  For a unit operated 500 hours per year, the baseline cost would
be approximately 86 mills per kilowatt hour.  In the extreme case of baseloading
a simple cycle gas turbine, the cost would be approximately 26.4 mills per
kilowatt.  Capital charges and fuel costs in the 500 to 2,000 hour operating
range are equally important.
7.3.2.3 Sensitivity Analysis - The baseline cost estimates are quite sensitive
to changes in the operating parameters.  In order to test the different sensi-
tivity, estimates were prepared assuming values for the underlying parameters.
These estimates VTCTB used to provide bounds around the baseline estimates
developed above.  Table 7.18 shows the resulting cost estimates and the
assumed values for the operating parameters.
       Although standby units are generally more sensitive to changes in
capital costs and fixed capital charges, at higher operating hours fuel costs
                                                      I
bee'1*  -• more important.  The industrial examples are quite sensitive to changes
in furl prices.  At the controlled interstate gas price of $.66/MMBTU, costs
                                    7-50

    -------
                                 TABLE 7.18
                    BASELINE COST SENSITIVITY ANALYSIS
                             Standby Unit

                              80 hr/year

                          Mills/Kwh      %
                               200 hr/year

                             Mills/Kwh      %
Baseline Estimate
Fuel
     .   $.66/MMBTU

High Capital Cost
     .   270/Kw

Fixed Capital Charge
     .25%
     .15%
Range
611.3

587.7


742.5
748.8
473.8
473.8 to
748.8
 -3.9
 21.5
 22.5
-22.5
-22.3 tc
 22.5
264.8

241.2


317.3
319.8
209'. 8
209.8 tc
319.8
   -8.9
.  .19.8
   20.8
  -20.a
  -20.0 to
   20.8
                             Industrial Anolirat.ion
Baseline Estimate
Fuel
     .   $.66/MMBTU

High Capital Cost
     .   150/Kw

Fixed Capital Charge
     .  25%
     .  15%

Range
  2000 hr/year

Mills/Kwh       %
 43.8

 23.7          -45.9


 47.5            8.4
                                                        8000 hr/year
 32.5

 12.5


 33.4
                              -61.5
                                2.8
46.8
40.8
23.7 to
47.5
6.8
-6.8
-45.9 to
8.4
33.3
31.3
2.5 to
33.4
2:5-
-2.5
-61.5 to
2.8

    -------
Raseline Estimate

Fuel
     .   $.66/MMBTU

High Cap.cal Cost
     .   $180/Kw

Fixed Capital Charge

     .   25%
     .  15%
Range
       TABLE 7.1R
       CONTINUED

   Utility Application

   500 hr/year
Mills/Kwh      %
 85.5
 69.8
 98.1
100.5
 70.5
 69.8 to
100.5
-18.4
 14.7
 17.5
-17.5
-18.4 to
 14.7
                2000 hr/year
               Mills/Kwh  •    %

                38.2
22.5
41.4
42.0
34.5

22.5 to
4?.0
-41.1
  8.4
  9.9
 -9.9

 •41.1 to
  9.9
             7-52

    -------
per kilowatt hour decrease by almost two-thirds.  The Industrial unit was
much less sensitive to changes 1n capital costs.  For example, a 25 percent
Increase in the costs per kilowatt only changed the cost per kilowatt hour
by 2.4 percent for the single shift operation and by 2.8 percent for the
baseload application.  The results on the utility side are mixed.  If the
unit is operated solely as a peaking plant, all five changes appear to have
the same relative impact, varying between plus or minus 20 percent.   If the
unit is operated more as a shoulder peaking plant (intermediate load at
approximately 2,000 hours per year), changes in fuel costs dominate.
        In  summary,  depending  on  the  assumptions  regarding operating para-
meters,  the  costs  per  kilowatt hour  may  range  anywhere  from 611.3 mills per
kilowatt hour  for  the  standby unit operating 80  hours per year,  to  12.5
mills  per  kilowatt  hour  for the  baseload  industrial unit operating 8,000
hours  per  year and  consuming  natural gas  at controlled  Interstate prices.
Table  7.19 shows the estimates that  will  be used throughout the  remainder
of the analysis.  These  estimates are believed to represent the  actual
costs  incurred  by  users  of gas turbines.
7.3.3  Impact of Environmental Control Costs
       Section  7.2 of this analysis considered various water purification
systems that could be used to provide water of the required quality and volume
to control N0y  emissions from gas turbines.  In this subsection, the costs of
            • A
these  water  purification systems are converted Into costs per kilowatt hour of
                                                      i
net generation  to determine the potential impact on the model plant applications
described  in Section 7.1.  These costs are compared to the baseline estimates
presented  in the previous section to determine the relative impact in each
application.
                                    7-53

    -------
                                TABLE  7.19


                           BASELINE COST ESTIMATES

                             USED  TO COMPARE  THE

                 INCREMENTAL COST  OF ENVIRONMENTAL  CONTROLS


                                 .(Mills/Kwh)


Application                           Baseline               Range
     a.   Standby Unit
         1.   80 hours                 611.3                  473.8 -  748.8
         2.   200 hours                264.8                  209.8 -  319.8

     b.   Industrial Application
         1.   Single Shift (2000 hr)    43.8                   23.7 -   47.5
         2.   Baseload (8000 hr)        32.5                   12.5 -   33.4

     c.   Utility
         1.   Peaking (500 hr)         85.5                    69.8 -  100.5
         2.   Shoulder Peaking         38.2                    22.5 -   42.0
             (2000 hr)
                                  .  7-54

    -------
7.3.3.1 Parameter Values -- The most important determinants of the cost of
using water injection to control NO  emissions from gas turbines are:
                               •    rt
        (1) the capital costs of the water purification
            system;
        (2) operating costs of the purification system;
        (3) the water injection rate;
        (4) cost of water transporting to the turbine
            site, if applicable; and
        (5) any efficiency penalties associated with the
            injection of water.
Each of these factors are addressed below.
7.3.3.1.1 Capital and Operating Coasts -- Section 7.2 considered alternative
water purification systems.  The systems selected for use in each application
were designed to meet manufacturers' required water quality specification
and flow rates.  Sized for normal operating practice, the units should ade-
quately meet the NO  emission levels considered (i.e., 75 ppm, etc.) proposed
                   A
by EPA.  A summary of capital and operating costs of the systems selected for
evaluation is provided in Table 7.70.  Capital costs were based on manufacturers'
estimates of installed costs and the cost of modifying the gas turbine to per-
mit water injection.  Operating costs included chemicals and other variable
expenditures for proper operation of the purification system.
       The same purification system would be adequate for either of the
                                                     i
standby applications.  If the turbine could be expected to operate more than
200 hours per year, a more expensive purification system could be purchased
which would substantially reduce operating costs.  Given normal operations
                                    7-55

    -------
                                 TABLE 7.20
                         CAPITAL AND OPERATING COST
                ASSOCIATED WITH WATER PURIFICATION EQUIPMENT
Application
A.  Standby unit

    1.  350 HP

    2.  1100 HP

B.  Industrial

    1.  4000 HP - typical

    2.  4000 HP - offshore


C.  Utility
 \

    1.  66000 KW
  Installed Capital
                                                       1
                                      Total $
170000
               $/Kw
2.58
            Operating  Costs
            $/1000gal    $/yr
1400
1400
14000
27000
5.46
1.73
4.78
9.21
20.00
20.00
3.00
11.00
9.00
22.50
175.00
9640.00
2.90    7300.00
   Includes both the purchase price of the water purification equipment and
   the installation cost.
                                    7-FG

    -------
for standby units, however, it 1s unlikely that a more sophisticated sys-
tem would prove to be economical.  The cost estimates for the industrial
applications are based on estimates provided by vendors of water treat-
ment equipment.     The offshore facility has considerably higher costs
because of the need to desalinize the injection water.  The utility esti-
                                                                     727
mates are based on a unit currently in service using water injection.
The system is designed to meet state emission regulations of 75 ppm (ref.
15$ oxygen) or the equivalent of regulations being considered by EPA.
7.3.3.1.2  Water  Costs -- With the exception of the  offshore drilling platform,
all of the water  purification systems are expected to  utilize tap water  supplied
by  local utilities.  Normal water charges per  1,000  gallons range from $.20 to
     737
$.70.  '  Table 7.21 gives water  costs per 1,000 gallons  for fifteen cities.  The
average cost of $.37 per 1,000 gallons was  used to estimate water costs.
 7.3.3.1.3  Transport Costs -- In a few isolated applications,  such as remote
 pipeline compression stations,  feed water for the injection system will  not
 be available at the plant site and will  have to be transported by truck  and
 stored on site.   The capital  cost figures in Table 7.20 include the cost of
 water storage facilities.   The costs of transporting the water have been
 estimated to be $.02 per gallon for a 50-mile, one-way trip and $.04 per
 gallon for a 100-mile trip.     It is highly unlikely that a source of
 potable water of sufficient quality would be more than 100 miles from any
 turbine location.  One major interstate gas pipeline company has stated  that
                                                           75
 water will be available at all  of its compressor stations.
                                      7-57

    -------
                                 TABLE 7.21
                           ESTIMATED AVERAGE COST
                            PER THOUSAND GALLONS
                              OF WATER IN 1974
City
    Dec.  '74
$ per 1000' gals.
 Out of City '
Limits Market
Albuqut-que, New Mexico
Atlanta,  Georgia
Austin, Texas
R31timorp. Maryland
Boston, Massachusetts
Detroit,  Michigan
El 'Paso,  Texas
Kansas City, Missouri
Long Beach, California
Louisville, Kentucky
Miami, Florida
New York, New York
Oklahoma City, Oklahoma
Seattle,  Washington
             Average
    .233
    .526
    .486
    .646
    .197
    .240
    .324
    .312
    .553
    .244
    ,704
    .282
    .155
    .366
    100%
     70*
     50%
    None
    37-40%
    $5.00 month
      8%
     50%
    None
    None
    None
     34%
     50%
Source:   Dallas Rate Survey, Journal AWWA, May '75.
   For users consuming more than 100,000 cubic feet per year.
   Potential increase in cost if located outside city limits.
                                  • •7-58

    -------
7.3.3.1.4  Efficiency Penalty -- One manufacturer has estimated that water
injection will result in a 2.9 percent heat rate penalty (BTU/kwh) and a 3.9
                                            76/
percent increase in the capacity of the unit.      Westinghouse estimates that
water injection will result in a penalty in the range of 1 to 1.5 percent with
a 3 to 4 percent increase in the capacity of the unit at baseload. '   In
evaluating the cost of the emission control system, a 1.5 percent heat rate
penalty was assumed to occur in all applications.  Enhancement of output
capacity was only considered when the unit would be continuously operated
at its rated capacity.  In such instances the enhanced output of the unit
might increase the annual  output of the unit, thereby reducing the annualized
cost per unit of output by spreading the fixed cost over a greater output.
However, this will  probably not occur in most instances because the purchaser
will install a turbine which is slightly larger than required to meet maximum
requirements.  The only exception would be for baseload internal power
generation and possibly compressor stations where turbines are operated at
full load between major overhauls.  The capacity expansion factor was in-
corporated by decreasing the installed costs per kilowatt by the percentage
increase in the capacity of the unit.  For example, if the cost per kilowatt
was $120/kw, the cost after adjusting for the enhanced capacity would be $116.50
per kilowatt, which is a 3 percent decrease in the installed cost.  All applica-
tions where this factor was considered to be important were analyzed with and
without this adjustment to ensure that the range of possible control  costs
are presented.
7.3.3.2  Cost Estimates -- The model plant control  cost estimates shown in
Table 7.22  were used to calculate the percentage impact shown in Table 7.23
Eleven model plant applications were analyzed.  The results in each case are
explained below.
                                        7-59

    -------


































CM
CM
ts."
ut
_J
»-
































1
tA
S




i"
O, ••-
•*••
§











g
U
E c
i °

°"-~-
o o"C


§ -g
«/> 3 r
Uj £
g B

£ §
m u
S £
2C S-
S
«t ui
£ 0
§
5
 i/

2=
^5



**







n
4-»
c
5




«
t.
OJ


ex
JC
o o o o i
g CM 0 CM
r— CM
* J2
s
J| S 8 8 '
-J8 2
s
O O O O r
o in o CM
f lO r— <"1
=>» 0
vo •-•
3
§O O O «-
m o c\j
Sg 2
4
I S § s '
1 40 O
3 U> 1-4

O
J= 0
8O Q  i- a:
CO O
»- CM S i 10 mo
A
CO
00 CO 53 ^ ,
*" Wl Ch fO 1 |A <"*>
-.~ | ~ N ' y \ ~ «
co
eo 0} o r«.
"~> , °l n. ! **! '

CM" 1

oo • co o r-~ i i
* • • • r "J I'
o ^^
,
co eo o r- i t
*-< in en ro i m t
" ~ g ~ ~ 3 ' ~ ' ,

eo co o r-«
Or- r^ o coioomi

mcMO*^ ro r— CM p— i
g
•
«o
co oo o r-. i
o»- r» o roioiino
•ncMO<- co- r— ir-co


a*

co co c^ r^ i i

in CM o » «v> ' ,J ! |J i

o
CM

o 2 C 8 co S • * i
i— CM O r^ O 1 r-^ 1
S '


1 1
co ro o r* m i i
Or— f*. o «*> CM i m i
- CM' 0 r.' 0 •-._••
CO CM



OO «O O f^
0 ™ .? 5 co »- ( in
r— CM O m O lr*l
O CM II
CM



CO «O O I*.
o ^-< «r o — *o o o *-x k
S S o o — p S2
^•^^ O CD K M
01 . *-> O C l« C ct JJ «J U
2 ° «.JJ c «- S c
CT ^ ° '" tl U- S 5"
C vt ^* *^ C H* O (U t c? >t
•^ O O •"• ' '^ *" '" O CJ£ K~ Q> 4^
±JCJ (O1.4JI. .^.^-
l/t
^
„
S
,
ra
:S



•
CO


8

tn

CM



in
CM





CO
in
w



co
5
CM



"I
1C*


CC

CM



in
^'
1C


•g
w
vt

r~
X




£
o
u

c

--a.  v- c,   t)  ai  oj.^
o    o.   ^  «o  o ^-
                                 7-60

    -------
                               l/»r- *->
                               «t- 'i-  10
                               «*-*-•-
                               OO O.
                                                                            CM
                                                                            CO
                                                                                  CM
                                             VO
                                             O
                                                         O  co
                                                                            cn
                                                                            CO
                                                                                     to
                                                                                     CM
                                                                                  in


f


CM
1


«-»
1
rs
VO r* f— • *^* 1 1
CM r- O CO 1 1
. . . . 1 |

CM
•


fft r- r- «d- 1 1
Or- 0 CO 1 1
. • . . l |

0>
•^"

*"~

CO r-H ,-1 «tf- 1 |
IO •-• O CO 1 1
... . | |
CM
s

CO
o
CM
•
00
co
m
*d"

in
00
o
o

o
                                                                                            in
                                                                                            vo
        I

l/l
3

CO
ri


CM
1
*-«

CM
r~



CM
r—
.


O




o
F—
•


i— CO *J-
o «* vo



f— CO 1
0 ^ 1



en
O
i*


en
0
.
i

^
CM
,_!


in
•


CO
in
CM
co

CO
in
•
CM
CO
                                                                                            co
                                              CO  O
            t—   CO
            o   «*
                                                                                CM
                                                                                o
                                                                                   CM
                                                                                   CO
                                                                                r-    CO    CM

^J-
1
to


CO
1
IO


CM
t
40



r—
1
I/)

CO

•
r-»

CNJ
CO
•
«*

VD

•
in

in
vo
*
CO
r—

^
CO
•


r-.
CO
,


IO

•


VO

, «



t
o
a


r—
O
•


r—
0
*


r1"
0
•



n*
in
.


_
in
•


^_
in
»


r—
in
•













































CM
VO
•
CM

r-.
CM
*
in

^.
^J-
»
vo

CO
VD
•


Ot
fs^
•
^f
CM
o*
cw
•
r~t
to
c^
f%«
»
vS
CM
en
CM

r*f
f*«
VO
0»
C7>
t


in
CO
•


CO

•
CM

O>
CO
*
CM


                                       V)
                                       O
                                      O


                                       O
                                      •I—
                                      4->
                                       O
                                       w
5   *.     .i
•O   O     -»J
 CJ      4J   U       *->
 X   4J C   O)   >,  t-
•r-   I/I O)  ••-»  •«-»   O
U.   O E   C  •—   CL
     •j *j  i—•   
                                                                O>  J- 10
                                                                V.   O) *->
                                                                
                                                                   E
                                                                   QJ
                                                                   O
                                                                   C
 C  O
UJ  »-


 a.

i


to*
                                                                                    CX
                                                                                    2:
                                                     7-61

    -------
7.3.3.2.1   Standby Units -- Four model  plant  applications were considered for
standby power generating stations.   The first two  cases, S-l  and  S-2,  differ
only in terms of th.  number of hours operated each year.  The first  unit
operates approximately 80 hours per year and  the second 200  hours per  year.
These units show the  highest percentage impact in  terms of the incremental
costs per net kilowatt hour of power generation.   The  low number  of  hours
operated each year tends to drive the cost of producing power up  because
fixed costs are spread over a relatively small base.   The estimated  impact
in both cases was roughly 2.4 percent.
           Cases S-3  and S-4 apply to 1,100 horsepower (hp)  units operating
the same number of hours, respectively, as the smaller 350  hp unit,.  This
units can use exactly the same water purification  system as  the  smaller unit.
Consequently, while the costs of producing power,  independent of  the water
injection system, are exactly the same, the percentage impact is  much  less,
decreasing to less than 1.0 percent.
7.3.3.2.2  Industrial Units -- Four model plants were considered  in  the in-
dustrial category.  The highest percentage impact  was recorded  in Case 1-3,
which represents a remote turbine application in an arid climate  in  which
water must be transported fifty miles at a cost of $.02 per gallon.   The
impact  in such cases, which includes water storage facilities,  is approxi-
mately  a 3.7 percent increase in the average cost  of generating  power.  Since
water  injection  results  in  a  slight  increase  in the power output capacity
of  the  unit,  a  credit was  taken of  .05 mills per kilowatt for the output
enhancement.  The  unit  is assumed  to be baseload,  operating  8,000 hours per
year.
                                        ' 7-62

    -------
       The remaining applications are as follows.  Case 1-1 represents a
normal, single shift gas turbine application.  The unit is operated 2,000 hours
per year and is slightly oversized, therefore negating any benefits that might
be derived from the improved output of the unit.  Case 1-2 is a baseload
turbine operated 8,000 hours per year.  A credit was taken, in this instance,
for the improved capacity of the unit.  All three units were sized at 4,000 hp.
7.3.3.2.3  Utility Applications — Four model plant applications were con-
sidered for the utility industry.  The first unit is operated 200 hours per
year, the second 500 hours per year, third 2,000 hours per year and the
fourth  8,000 hours.  A credit for enhanced output was taken in the last case,
since the unit is baseloaded.  In all four cases, the percentage impact per
kilowatt hour is less than 2 percent.
7.3.3.2.4  Offshore Drilling Platform ~ Gas turbines are commonly used on
offshore drilling platforms.  Their compact size and low maintenance require-
ments favor their use for this type of application.  Initially, it was/thought
that this case would evidence the highest percentage impact.  The unit was
assumed to use sea water to fuel the water purification system, resulting
in a substantial increase in the capital and operating cost of the purifica-
tion system.  The installed cost of the combination desalination - purifica-
tion system was $27,000 compared with $14,000 for the onshore application.
Despite these high costs the fact that water would be available more than
offset the higher costs associated with transporting water to the remote
gas compressing station application.  The total cost for water injection at
the remote site was 1.21 mills/kwh compared with .92 mills/kwh for the
offshore platform.
                                    7-63

    -------
7.3.3.3  Sensitivity Analysis of Cost Impacts — The cost estimates  discussed
above are partially dependent on assumptions underlying the cost model,   Not
every firm or utility \.ill  consider a fixed charge rate of 20 percent to be
appropriate and, regardless of its availability, some firms will  use controlled
interstate natural gas in new turbine applications.   Recognizing that the
assumptions made in the baseline case will  not hold in all cases, four of
the parareters were given different values  in an attempt to determine the
sensitivity of the cost estimates.  Fuel  costs were reduced from $2.18/MMBTU
to $.66/MMBTU, which is equivalent to the current city-gate price for natural
gas.  The fixed capital charge was given values of .25 and .15.   Capital
costs were increased to $270/kw fortstandby generators, $150/kw for industrial
units, and $180/kw for the 66,000 kilowatt utility turbine.  Finally, the
output enhancement adjustment was dropped,  eliminating the credit for the boost
in power output associated with water injection.  The resulting percentage
impacts after making these changes are shown in Table 7.24.
           The estimates are most sensitive to the change in fuel costs.
In eight of the eleven model plant applications, reducing fuel costs from
$2.18 to $.66/MMBTU resulted in the highest percentage impacts.  The change
was especially  pronounced  in the  two cases where high water purification
costs were anticipated.  Lowering fuel costs in the remote site industrial
application (1-3) accentuates the importance of water transport costs.  In
this case even  though environmental control costs decrease, the decrease in
the cost of power generation was  even greater, thereby resulting in a net
increase in the percentage impact.
                                         7-64

    -------







4
ENVIRONMENTAL
CM
• U-
r» o
UJ OO
_J t— «
9 52
H- _!
B
2»
1—
r-H
l/H
Z?
UJ
M








e?a
££-£
 r- 4-
*»»•••«;
U- t. i-
ooa
«»•
i
3
.*«?
^ i
+J
=>
CM
1
=3
ft
1
uT
fc 7
o *-•
o
£ LU
£ * .
s 2 .
_«9
O < "C CM
2 CO *> 1
P trt >-<
So 5
8 c
** •"•
' T
I-*

CO
1
t/l
s
•o •
s
«/> CM
1
CO

r™
1
«/]
s .§? s
CM W CM

to in to
in to «r
§«t r*.
"". m.
f— CM f-^
in o »j-
r«. oo r>.
•~ •" -•
o> to r-»
to m en
• * •
«-l r-< r-4
•
r- Ol f.
r<. to GO
CO . O> CO

u> *»• co
f«. t— !••
» » •
r- CM •—
CM 10 CO
CO CO <*
CM CM CM
a 35 «
• • «
S S S
* • •
f-

CO CM *
* «". 5.
CM CM CM

O* CO r«"
CO «* *f
CM . CM CM
I 1
*> S -c
u +•»£<_>
•O Vk ^.
to oo f
o r-  « *»
4* •— CM -»-
+» N > n n>
O r- U- « O «
o  in
. e u -o CM
at < a? to a»
c S to x
•t- i. O • *r*
T 9 — ' *» "-
C 00
C*J r^ r—
*>• »«l CO
s s
« CM
.
& 5 a i
^ ,:.!
0 ,* I
o» rv.
*"* ^~ 1
U> CM
^ „,
i
CM lO
r-i ft
• •
CM - r*
co «f I
t-> uS
• •
CO 'CO 1
•
!r~. co
to «*
• •
!»• »^
«»• *
•— r»
• •
CM r-
!0 CM
ao in
• •
r- r—
ICM ««
»-l •*
• •
CM **
o> ^
tb in
• •
O» CO
"
10 CO CM
f» 10 0
• • •
r- r- ' CM
0 «•
CM 7
• •
CM CM
S
^ CO
5 °
*
2 S
CM* -•
Is *
Co r-
• •
CM • CM
IO CM
CM . 00
-• •
ItO O
S i^
~
-
S S
• • 3
CM CM S.
IOV CO
at o
• •
CM CM
3'
SB; -8
CO P! ' g
I 8. ^
1 CO r^
4J 5
9 3 ,. .. fc
3 €-5335
r- O -^ ~^. »•,
5>>r« U O >>Q 4*
JO CM 4-> U> <-•» C
••- T>V> IA r- •^•- Ot
a. c i 3 i •— i 9
•o 
    -------
       The assumptions pertaining to capital costs and output enhancement, in
general, did riot cause any major changes in the percentage impacts.   Increasing
the cost of turbines relative to the cost of the emission control  systems reduces
the percentage impact, which was anticipated.   The fixed charge rate generally has
less of an impact, because it is applied to both the capital  cost of turbines
and the control equipment.
       In summary, the cost of water injection control equipment for gas turbines
should fall somewhere in the range of 2 to 14 mills per kilowatt hour for standby
units, 6 to 1.2 mills per kilowatt hour for industrial applications, and 4 to
3.0 mills per kilowatt hour for utility gas turbines (Table 7.23).  On a per-
centage basis, the potential impacts for standby units fall within a range of
.7 to 3.0 percent.  The range for industrial units is 1.6 to 9.7 percent and
for utilities 1.4 to 2.2 percent (Table 7.24).  Typical impacts will probably
fall in the range of 2.0 percent regardless of the application, or rougnly
1 mill/kwh.                                                            /
7.3.4  Impact of Environmental Control Costs for Regenerative and Combined Cycle
       Gas Turbines
       No attempt has been made to provide detailed estimates of the control
costs for  regenerative and combined cycle gas turbines.  The impacts, in abso-
lute terms, are not expected to be much greater than for simple cycles moreover,
the percentage impacts will be less given the higher cost per kilowatt of
generating capacity for these units.
       For combined cycles, the control costs should be the same as for simple
cycles.  If the costs of the associated boilers and steam generators are
included, the percentage impact is certain to be less than for the simple cycle unit.
                                    7-66

    -------
                                                         78
       Manufacturers report that to control NO  emissions   the regenerative
                                              A
unit will  require a h'Igher water injection rate than for similar s1ze» sim-
ple cycle  turbines.  Therefore, the purification system will  have to be
slightly larger and cost more.   For example, a 20 percent cost Increase for
a 4,000 horsepower unit would increase the cost by .03 mills  per kilowatt
hour for a unit operated 2,000 hours per year, the equivalent of a .1 per-
cent increase in overall operating costs.  If the water for the purification
system also had to be transported, a 20 percent increase in the volume trans-
ported would increase the cost by .13 mill, or 4 percent.  The maximum impact
in this extreme case would be .5 percent over the estimate impact of 3.77
percent.  Given these circumstances, it is believed that the  cost impact for
combined cycle and regenerative gas turbines will not be substantially greater
than for the simple cycle, and the percentage impact should be in the same
relative range.  In the case of the combined cycle, if the entire turbine
steam generator package was considered, the percentage impact would be smaller.
7.3.5  Alternative Control Levels
       All alternative control  levels less stringent than the 0.3 pounds per
MMBTU under consideration by EPA would require less control.   If the levels were
increased to .5 pounds per MMBTU, or 125 ppm, the smaller standby units would
probably not have to use water injection.  The cost in this instance would be
zero.  Also, for the larger units the water injection rate would be less,
reducing the amount of water consumed per kilowatt hour of net generation.
                                    7-67

    -------
However, control costs would not decrease substantially because the water
Injection ratios wo'Jld be much  less Important to the overall  cost of control
than the purity of the water and associated purification equipment.  The
cost per kwh for injection water is roughly .01 mill in every application.
 If  the  required water injection  rate  fell by  90 percent,  it would  only
 redt ^  the  cost of the system  by  .009 mills/kwh, which  is equal  to .03
percent of  the  cost of operating the baseload utility application.
7.3.6  Summary
       This section has  attempted  to provide estimates of the potential
cost impacts of NO  control equipment for gas turbines.  The resulting
 V                 •*
estimates showed that, except  for  standby units, the total change  in costs
will probably fall in the  range of .4 to 1.5 mills  per kilowatt hour for
turbines used in industrial and utility applications.  Control costs for
standby  units,  on a mills  per  kilowatt hour basis,  were much higher,
 ranging  from 2  to 14  mills per kilowatt hour, but this  is primarily a
function of the low number of  operating hours.  These costs are equivalent
 to  about a  2.5  percent increase  in operating  costs  for  standby units and
 about 2  percent for industrial and utility applications.
                                  7-68

    -------
7.4  ECONOMIC IMPACT
7.4.1  Impact on Manufacturers
       The economic impact of imposing NOX  emission standards on the gas
turbine industry will  depend on the industry's ability to pass through the
additional costs to consumers.  Table 7.25  shows the expected increase in
installed cost that will  result from using water injection to control NOX to
75 PPM.   The percentage impact varies anywhere from .8 percent in the case of
the 1100 hp standby unit  to 7.1  percent for the unit requiring both desal-
ination and purification  equipment.
       The  major questions  that must  be resolved are  the  degree of competition
between  individual manufacturers and  whether  the pollution control cost will
place  the gas turbine  industry at a  competitive disadvantage  with  stationary
reciprocating engines.   Regarding the first issue,  since  all  manufacturers
would  be  forced to use basically the  same purification  system to meet the
75 ppm emission level, no  manufacturer would  gain  a competitive  advantage.
Therefore,  the  degree  of intra-industry competition should remain  fairly
constant.
       The second and moro  important  issue is  the competitive  position of  the
 industry relative  to stationary diesel generators.   With  respect to  the  largest
 segment of the  industry  (measured  by capacity),  utility peak  load generators,
diesel engines  can not be  produced in this size range.  Therefore  it can  safely
be assumed  cross-price substitution  impacts can be  ignored.   The same can not
be said  for standby generators and gas turbines used  by the industrial sector.
In both  of  these areas the  competition between turbines and diesel engines  is
quite  strong.
                                     7-69

    -------
                                 TABLE 7.25
                       IMPACT OF NO  EMISSION CONTROL
                                   rt
                ON THE INSTALLED CAPITAL COST OF GAS TURBINES
Application
  Installed Cost (1000$)
                    % Increase
                                without
                                control s
                   with
                controls
A.  Standby
    1.  350 HP
    2.  1100 HP
  56.6
 177.9
   58.0
  179.3
2.4
0.8
B.  Industrial
    1.  4000 HP - typical      352.8
    2.  4000 HP - offshore     352.8
               366.8
               379.8
                    4.0
                    7.1
C.  Utility
    1.  66,000 Kw
9900.0
10070.0
1.7
                                       7-70

    -------
       Historically, the price competition between the gas turbine and diesel
engine industries has centered around first cost and fuel efficiency tradeoffs.
The cost of reciprocating engines varies anywhere from 30 to 60 percent more
than turbines, but they have the advantage of requiring less fuel per horse-
power hour.  While fuel efficiency depends on the type of operation, diesel
engines are generally about 30 percent more efficient.  Typical heat rates
for turbines run in the range of 9700 BTU/hp-hr compared to 7,000 BTU/hp-hr
for diesel engines.
                               t
       The cost of generating power from these systems is plotted as a function
of annual operating hours in Figure 7.2 for both the standby and industrial
application.  The assumptions underlining these curves are:
       (1) the diesel is 30 percent more efficient that
           the turbine;
       (2) the installed cost of the diesel is 30 percent
           more than the turbine;
       (3) both are operated on distillate fuel oil priced at
           $2.18/MMBTU; and
       (4) operating and maintenance cost are essentially the same
           and both include the cost of environmental controls
Given these assumptions the curves show that the diesel is clearly superior,
from a cost standpoint, to the turbine in industrial applications where the
unit is operated 2,000 or more hours per year.  The opposite is true in the
case of the standby unit.  The smallest cost differential for the industrial
application is 5.1 mills/kwh, which is slightly less than five times greater
than the estimated cost of the water injection emission control system of
1.02 mills/kwh.
                                    7-71

    -------
1500
                                           STANDBY>/yr
                                                120
1200
                                                           O DIESEL STANDBY
                                                           Q TURBINE STANDBY
                                                           A INDUSTRIAL TURBINE!
                                                           O INDUSTRIAL DIESEL
                                                                                                 &
                                                                                                 3
                                                                                                 O
   2000
                              BOOB,
                         , INDUSTRIAL, hr/yr>v ,  WM
Figure 7.2. Comparative cost for gas turbines and diesel generators."
1000
                                                7-72

    -------
The difference is more pronounced in the case of the standby unit where
the smallest cost differential is 52.2 mills/kwh compared to the
incremental cost for emission controls of 2:62 to 6.44 mills/kwh (Table 7.23).
      Another way of focusing on this issue is to solve for the number of
hours of operation that would equate the two systems.  The formula below
can be used to determine the breakeven point:
      Mil1s/kwh= (CRF)  (K) + (OH)             ..
                       Hr/yr      ••        pf  (L)

where
       CRF = capital recovery factor = 20%
         K = installed cost of unit (diesel  = 1.3 turbine)
        OM = operating and maintenance expense (assume .05 K)
        pf = price  of fuel in $/MMBTU (assume $2.18/MMBTU)
         E = heat rate efficiency in MMBTU/kwh (assume turbine =1.3 diesel)
To solve for the number of hours of operation that v/ould equate the two systems,
values must be assigned to the parameters.   The assumed values for
each of the parameters and applications are shown below.

                                Assumed Parameter
                                   Values
                                                  Turbine          Diesel
Standby
       CRF (%)                                       .2               .2
       K ($/kwh)                                    220              286
       OM (% K)                                     -ObKT            .Q5KD
       P.f(WMMBTU)                                 2.18             2.18
       E (MMBTU/kwh)                                .0155            .0109
                                        7-73

    -------
Industrial
       CRF  (X)                                       .2                .2
         K   $/kwh)                                  120               156
        OM   % K)                                    .05KT             .05KD
        Pf   $/MMBTU)           -                     2.18             2.18
        E   MMBTU/kwM                               .0132             .0092
The breakeven point for the standby unit is 1,650 hours per year; for
the inJustrial  application it  is 1,034 hours per year.  If these units were
operated less hours per year,  the turbine would have a competitive advantage,
the opposite being  true if the hours of operation exceeded these limits.
It is  important to  note that standby units are seldom if ever operated
more  than  200 hours per year and industrial turbines are normally operated
a minimum  of 2,000  hours per year.  If the sole determinant of market share
between  turbines  and diesel engines were the cost of operating the unit,
turbines would  have total control of the standby market and diesels would
control  the  industrial market.  While cost is certainly important.it is
not the  sole determinant of the end user's selection criteria.  Size, weight,
and dependability  in many cases are equally important and in some instances
the overriding  consideration.
       The  future growth and profitability of the gas turbine  industry will
depend upon many factors other than pollution control costs.  The most
important  factor is probably the price of fuel oil.  In the last section the
cost  of controlling NO  through the use of water injection was shown to  fall
                      ^
in the range  of .5  to 1.2 mills/kwh.  This cost impact  is approximately
equivalent to a 1.7 to 4.2 percent increase in the  price of distillate fuel
oil.   If fuel oil prices continue to increase as rapidly as they have over
the past few years, in the absence of substantial fuel efficiency gains
any competitive advantage once held by the turbine  industry over diesel
will  be  rapidly eliminated.
                                      •7-74

    -------
       On the basis of the proceeding discussion it would be difficult to



make a case for NO  controls having a substantial economic impact on the
                  A


manufacturers of gas turbines. . First, based on existing emission rates all



manufacturers will have to employ the use of wet control to meet the 75 PPM


NO  standard.  This would imply that  intra-industry competition should remain



constant.  Second,  in comparing turbines with diesel generators first cost



and fuel efficiency tradeoffs are important but not necessarily the only



determinants  of  choice.   If  price alone was the  primary determinant, the



market  would  be  considerably more segmented than  it  is.  Third, the most



 important  market based  on capacity  for gas turbines  is  the electric



utility industry;  diesel  generators  are not produced in this size range.



Therefore  the imposition  of  NO  controls  should  not  impact on  either the
                              /\                          "


degree  of  intra- or inter-industry  competition.   Since  the market



 positions  of  manufacturers  in the industry will  not  be  affected by the



 imposition of NO  standards, the manufacturers should be in a  fairly
                 J\


 strong  position  to pass additional  costs  on to end-users.



 7.4.2  Impact on Consumers



 7.4.2.1 Utilities — The impact of controlling  NO   emissions  from gas turbines
            "*~ "                                    n


 on end-users  should be  primarily a  function of the number of operating hours and



 the user's dependency on  turbines for power generation. The most important



 market  for gas turbines is  the  electric  utility  industry, where turbines  account



 for  approximately 8.25 percent (39,300 MW) of total generating capacity and


                                     797
 1.6  percent  of  net power generation.     Total electricity production by the



 electric utility  industry over the twelve months ending in June, 1975 was 1,886



 x 109  kwh.-    The average cost of generating power was 27.65 mills per kwh.81/
                                     7-75

    -------
If all the electric utility industry's  turbines  were  using wet  controls,
the percentage impact on the cost of generating  power would  have  been
.09 percent, based on an average cost of 1.5 Mills/kwh for wet  controls
and 1.6 percent of net power generation attributable  to gas  turbines.  To
place this figure in perspective, a change in the cost of generating power
of this magnitude would be roughly equivalent to a .5 percent increase in
the price of coal.  Since the emission  level under consideration  here, if
adopted, would only require control on  new gas turbines, the full  .09 per-
cent increase in the cost of generating power will occur until  all  existing
turbines have been replaced.  Gas turbines have  an extremely long  operating
life (in excess of 20 years); therefore, the total impact will  not be felt
until somewhere in the 1995 timeframe.
       Also, since the utility industry is a franchised monopoly,  they should be
allowed to recover any costs associated with water injection to control  NO
emissions from gas turbines through rate adjustments.  This  has been true  for
other pollution control expenditures and there is no  reason  to  assume  differently
in this instance.
7.4.2.2 Industry — The impact on the industrial sector is a function  of each
industry's dependence on electricity and the degree that it  relies on  internal
power generation.  Table 7.26  shows the amount  of energy consumed per dollar
of output in the seven most energy intensive industries.  Also  shown  is  the
percentage of total energy consumed in the form of electricity  and the percent
derived from internal power generation.  The cost of wet controls on  gas turbines
can be converted into a cost per million BTU simply by dividing the incremental
cost for the NO  emission control system (1.2Mills/kwh) by  the  amount  of energy
               A
(.0132 MMBTU) consumed per kwh of power generation.   Therefore,  the  incremental
                                    7-76

    -------
                    c
                    O  -M
                         O
                    4J   3
                    O  "O
                    «O   O
                    Q.   V-
                    E  CL
                        tt)
                        to
                        «o
                        Ol

                        o
          O)
          U
                                                O  ro  00      vo
                                                OJ  r—  VO   O  VO  C~»   r—•
                                                O  O  O   O  O  O   O
                                                                                                    en
                                                                                                    OL
                                                                                                          V.

                                                                                                          0.;
                                                                                                     1C
                                                                                                     ft1
                        4J
                        c
                        o      *-   3
                        •r-   OL  U.
                        3
                        cr   c  s-
                        Ul   ••-   O
                                           Cn   CF>  O>  Cn   Ol  Cn  Cn
                                           O   CD  O  O   O  O  O
                                                                                      O)
                                                                                     •c—
                                                                                      V-
                                                                                     •o
                                                                            3
                                                                            C
                                                                            «o
                                                                                                    C/i
(JO
               to
               Ul
     CC   r>
     o   o
     u-   2:
          >—<
  x to
O   UJ   UJ
sr   o   >

U.   0£   -
     (—<   O  O
2=   _J   f-
UJ   O   CO
»—   Cii   O
o   i-   se
o.   -z:
     o   z
     C^   UJ

          Ul
          to
                                CO
                                 •o
                                  0)
                                 4->
                             «*-   ro  ^>
                             f-   i-  a*
                             Q)     csj   r-
                                           <£>  OJ  CTi
                                           r-  r-  C\J
                                                              o  r«-

                                                              O  r—
                             (O
                             U
 01

 «o
JC   r-   C
I/)   Ul   UJ
                                  en
                                  $-
                                               vo  eo   o  o  o
                                                    in   oo  en
                                                    r—   r—
                                                                                      8!
                                                                                      0>
                                                                                          &
                                                                            o.  =
                                                                            E
                                                                            3   X

                                                                            §    r
                                                                            O    i/)
                                                                                  c
                                                                            *»•    c
                                                                                          Q
                                                                                          cn
                                                                                          o
                                                                                          X
                                                                                          to
                                                                                          a
                                                                                          o
                                      CX
                                                                                                    (U
                                                                                               O    U
                                                                                               
                        o  CM
                        <$•  h-
                        •-"  O
                                                         o   a\  o
                                                         in   co  «a-
                                                         O   r—  r—
                                                                            vo
                                                                            f\l
                                  QJ
                                  o
                                      CO
                             VI
                             3
                            TJ
                             C
                                                    ro
                                                    u
                                                O)
                                 to
                                 u
                                 •r"

                                 a>

                                 u
                                 o
                                                              E

                                                              (U
                                                                   e
                                               «/>
                                                    -4J   4J   D.
                                                       (U
                                                    a.   a.  a.
                                                                       a>
                                                                           o
                                                                                     O)
                                                                                     o-
                                                                                          O)

                                                                                          Ul
 3   M
TJ   t-
 C      Ul  i-  «f_
                                                                                     O       4J   O
                                                                                     o.    «   a.
                                                                                          Sin   E   >»
                                                                                          l*»   3   <1>   r~
                                                                                     a.   cn   w   ->    .
                                                                                          f—   c   L   ^j-
                                                                                     «*-        031
                                                                                     O    «  O  00   *-»

                                                                                     •M   r-   >>  r-   <
                                                                                     O        O>   «O   >-'
                                                                                     «o   x:   w.   3
                                                                                     o.   u   
    -------
cost per kwh for NO  controls on gas turbines would be equivalent to approxi-
.  ^j                x
m?tely $.09/MMBTU.  By multiphying the energy/output coefficient for each of
these industries by t'j share of electricity, the percentage of electricity
which is self generated and the fuel price equivalent of the cost of wet
controls, it is possible to approximately indicate the potential impact on
product prices for the highly energy intensive industries.   The price irises
most in ine case of aluminum where, if all internal power was generated
from gas turbines using water injection, the imposition of a 75 ppm NO,
                                                                      4\
standard would lead to a .079 percent increase in product prices.  These
estimates are certain to overstate the price impact, because many sources of
power include hydro, steam electric and diesel engines.  The figures in
Table 7.26 indicate that even under worst case assumptions  the impact on
industrial end-users will be quite small.
7.4.2.3  Standby  -  The potential impact on users of standby turbines is
also believed to be quite small.  Since the units are only operated once
or twice a year and the capital cost is expected to rise by .8 to 2.4
percent, the economic impact on end-users of standby units  will be quite small.
7.4.3  Total Capital Cost,_Annyalized Cost and Energy Penalty
       The total cost to the nation of imposing a 75 ppm NO  standard on the
                                                           /\
stationary gas turbine industry has been calculated on the  basis of expected
sales for the period 1977 to 1982.  Sales projections were  based on Section 3.1
of this report.  Costs are shown both in terms of total capital outlays and
annualized cost as a function of assumed hours of operation.
7.4.3.1  Projections of New Capacity
        Table 7.27 shows turbine capacity on-line in 1974 and projections
                                       7-78

    -------
                                TABLE 7.27
                               •
                              SALES PROJFCTIONS   ,.
                          AND INCREMENTAL CAPACITY17
                                         Total  Capacity
Application                      1974	1980	1985
7/77-6/82
Utilities (MW)
    EEA (High)              •     39,300     57,000      77,000 20,972
    TBS (Low)                    39,300     50,700   .   64,000 13,676
          *
Oil and Gas Ind.  (103HP)

    A-D (High)                    6,831     10,755      14,904  4,509
    BAU (Low)                     6,831      8,242       9,423  1,429
Standby (103HP)                   1,257      2,450       3,952  1,509

Private Electric  (MW)
    Power Generation
    a.  oil/gas Ind.
        A-D (High)                  460     1,003        1,164    419
        BAU (Low)                   460       500     .     536      41

    b.  other industry            1,900     2,695        3,600    943
Other applications (103  HP)        1,147     -1,627        2,175  .  570
  I/ Sales projections  based on  data  presented in  Chapter 3.
                                       7-79

    -------
for 1980 to 1985.   To conform to the economic impact guidelines these estimates
IK;e been adjusted to a five year period beginning in July of 1977 and
ending in June of 1982   Incremental sales for this period are shown in the
last column.   All  cost estimates and energy penalties are keyed to these
sales pi ejections.
7.4.3.2  Capital  and Annualjzed Total  Costs  -  Table 7.28 presents estimates
of the t ,al  capital outlay  required to equip gas turbines with water injection
systems to control NO  emissions.  The total capital cost for the years 1977 -
                     /\
1982 should fall  somewhere in the range of $53 to $85 million dollars.  The
total annualized cost will scale up from zero in 1976 to between $19 and $40
by mid-year 1982.   Assuming  a linear growth rate the total cost over the five
year period should fall somewhere in the range of $60 to $120 million depending
on sales volumes.
7.4.3.3  Energy Penalty  -  The energy impact of these regulations will depend
on the efficiency loss associated with using water injection systems, the heat
rate (Btu/kwh), hours of operation and total capacity expected to come on-line.
Table 7.29 provides two projections that vary only in incremental sales volumes.
The energy impact is projected to be between 1,500 and 3,800 barrels of oil
per day by mid-year 1982, compared to a total nationwide consumption of 18
million barrels per day, or .02 percent.  This increase in fuel oil consumption
will gradually be reduced to zero as emission levels are met by the application
of the dry control techniques discussed in Chapter 4.  Most companies anticipate
that dry controls will be applied within five years.
                                       7-80

    -------
                      I— <">
                       "3 CO
                       3 01
                      r-   VO
                       10  4J O
                       o o
                      h- O
                      CM   i—

                      o   in
                                           r-»   CM
                                           in   vo
                                                          en
                                                          to
                      •o
                               ,
                       c:  o   ._.
                      <  o   ac
                                                    •M

                                                    in
                                                                           o
                       >- a)
                        n)
                          => ,
                       (1) r- -~»
                       K "3 i-
                       § 3 X
                       10 C"—
                      O  O  O
                      O  O  CO
                      U1  O
o  o
o  o
o  o
                                                      CM
       8
       a
s
 o
o
                      10 CM
                     4-> CO
                     •r- CTl.—.
                      CXf— 
                      10  I
                     o t-*o
                        r- o
                     r- CTl.-
                      CM  •—

                      in  in
                               CO

                               in
              o

              (M
                      r—  CM

                      s  «
O

CM
                  CM
                  in
IO
s
       o.
       s
       ?
       o
 c
 O  U)
                      QJ  Q.
                     *J  n>
                      10 <_>
                                           CO  CM  ID
                                           in  r~.  «j-
                                        oo  co

                                          •    •
                                        «*•  *J-
                                                                           co
                                                                   v>
                                                                   o
                                                                   a.
                                                                                        4J
                                                                                        in
                                                                                       I
                          s
                     CM   *J
                          c
                     ri   £  J=
                     r-.   u   
                     o»   c  f
                      10  CM
                      CM  f^
                      10  O
                                                                           cS
                                                                           «tr
                          1
                         r™
                         5
                          o
CM
O»
O
CM




it
a)

4->
•^—
•r-
4-1
3
CM
CO



«/>
3

•o

10

«r-
O







>,
JY
x>
c:
IO
4-»



I.
O
g
O.
»— •

^;
(.
4J
C

en
in
IO
01

^:
o

«
 r—
IO IO
U 4->
"S. *~
n.
10

s.
5

o
t
c
o
u
c
CJ

£

(.
s
o
(X

(/)
1.
o
re

                                                                                              7-81

    -------
 >,*•>
 o» o     »•
 I-  n3   J£
     r-
                               ^    10


                               CM    f-»
                                 *      M

                               00    CO
                  Or-
                  •r- 10 *
                  •t- c
                  <»-   rj  n>
                  •a;  v>  o>
                     ^>  >>
                  Tl    ~^
                  (II
                      -
£5
                         S~
                      3  3
                      c  o
                      c 1C
                 o
                 o
                 10
o   o
O   CO
o
  A

00
o  p
    So
    o
                                                     o
                                                       *

                                                     CM
                       jt
                       o
                   c
                   Ol
                   U   CJ>
                   C  1-
                  •-•  a:
                                   ID
                                   CM
                                   10
                          CM
                          r>.
                          o
                                   CO       r—
                                                                       00
                                                                       CM
                                   CM
                                   r-»
                                   CTI
                  o
                  CM
                          CVI
                          CO
                                                                                                    3


                                                                                                    o
                                                              •o
                                                               c
                                                          I/I  T-
                                                            (-
                                                       c
                                                       o
                   11
                   u
•a
 c
 10
                                >>
                               3
                               "cj
                                                          O   O       r-
                                                                        Q.
                                                                        ex
                                                                        M
                                                                                                    i-
                                                                                                    a»
                                                                                                    in
                                                                                                      •
                                                                                                    o
                                                          n)  .0
                                            o   y>  •-•
                                                                        0)
                                                                       J=
                                                                       w
                                                                       o
                                                                        7-82

    -------
7.4.4  Impact on Balance of Trade

       The export market for gas turbines has been growing quite
rapidly over the past five years. .The imposition of NO .  standards
                                                       ^t
should not affect this market potential because the controls on
NO  will be -external to the basic operation of gas turbines.  Unless
  J\i                      •
the foreign buyer required WO  controls, turbines available to
                             X
satisfy foreign demand will be essentially the same units that are
                                t
currently available.  If dry controls are developed it is likely
that the costs to meet the 75 PPM NO  limitation will be  no greater
                                    X
and probably less, both in terms of first costs and operating costs
to the consumer than the cost of water injection.  Any research and
development costs for dry control will probably be amortized over
total sales, including foreign sales which would affect the purchase
price to foreign buyers.  Given the status of research on dry con-
trols it is impossible to estimate what it will cost to develop
dry controls, which also make it impossible to estimate the poten-
tial secondary impact on e::port markets.

7.4.5 'Impact of Alternative Level of Control

       The alternative levels of control which will be considered
here are 100 and 125 PPM- (Ref 15% oxygen).  If'these standards were
adopted, a significant percentage of the gas turbine industry
would not have to use water injection systems to control  NO .
                                                           J^
Based on the results of Section 7.2 only the large turbines in
the range of 20,000 MW and above would be required to use water
injection systems. Table 7.30 shows the distribution of sales by
size range and manufacturer in this size range.
                                                       .   , »
       At present, for those machines which require water injection,
the only difference among the alternate levels of control is the
amount of water which must be injected into the turbine.   As dis-
cussed earlier, the costs of water injection are not that sensitive
to the amount of water injected over relatively small differences
in water flow rates.  The decrease in the flow rate should result
in a 10-15?, reduction in the capital cost of the water treatment

                            7-03

    -------
facilities.  This capital cost decrease would reduce overall
cost impacts from 1.6-1.9% under the 75 PPM level to 1.5 to 1.8%.
This would reduce annualized cost to utilities by $.7 to $1.1
million per year.

7.4,6  Cost Effectiveness of Control System

       Table 7.31 shows the cost-effectiveness ratios measured in
doll  :s per pound in NO  emissions.  The range for utility and
                       J^
industrial applications is from $.07/lb to $.25/lb of NO  emission
                                                        X
reduction.  The cost-effectiveness is considerably lower for stand-
by units, ranging from $.87/lb to $4.88/lb.  Standby units have
considerably lower NO  emissions characteristics which, when com-
                     *rt
bined with the high control costs, lead to a much higher cost per
pound decrease in NO  emissions.
                    X
                            7-84

    -------
 o>

 «3
JZ
to
 QJ
.X'
 t-
                                 CO
                       CO
                                          CO
                                            •
                                          in
                                                   «fr   CM
                                                            o
                                                            o
     3

     O
CT>
              IX
              3:
                       CT>
                       o
                       oo
                       CO
in

CO

en
CO

ui
"Kf
«*  i—
  •    »


CO  CO
                                                             oo
                                                             in
to ,/)
i £
13 ni
— fy
*"""" jCZ
o 5 *s>
^ ^
t^ "» 
N
•r-
to



^ CO
•K O
Or- D.
CD ** rrr
7\3




1 OJ
CVJ O
in r— o.
o CNJ a:
r>- co
IO r—




,
f^^ r~"
CM in
in o a.
CO h^ 1C
co to







o
«
o^
o

CO





«tf-
•
CO
o
r—


O
•
o^
oo
CO



•*
•
CO
o
en






r_
•
CO
f—
f—


^o
«
r—
CO
to
CO
CM
to



OJ ^t
• «
o «*
r^- oo
CO 3-
O
•
r-
to

m





O r~
• •
m CM
*t CO
r— r—

to
•
CO

«fr

        i
       in vo
       r— CS)
       r— in  Q-
       O CO  31
       CVJ CO
                        V£>
                        VO
                        cn
                                                             to
                                                                                     0>

                                                                                     D)
                                                                                     C
                                                                                    •f—
                                                                                    •o
                                                                                     C

                                                                                     o
                                                                                    •o
                                                    I
                                                    &
                                                    o
                                                    o
                                                                                                   CO

                                                                                                   C
                                                                                                   o
                                                                                                   o
                                                                                                   O)
                                                                                                   to
                                                         •o
                                                         (U
                                                         to
       J-
       0)


       4-J
       o
       (d
       <»-


       (O
                        O
                        •r*
                        i.
                        ••->
                        O
                        0)
                        (O

                        OJ

                        QJ
08

a.
          0)
          en

          o

          cn
 01
 CO)
 a>


T3   to
 o   s-
ja   
    -------























•-y
c5
CO
>-H
ct:
ct
D-
2:
O
0

t-l CO
CO CO
• UJ
r- z:
UJ
0) >•
f— t— *
.0 \-
0) O
^ £
U.-
U-
UJ
1
1—
CO
o
O






















(O
4J -t->^-.
'•r- IO -CO-
Q. O 	
<0 O
O
CO
CO s •*
O) X
C 0
OJ Z
>
•r- CO
4-> X)
! O r—
•»-> CD ~^~
co H— ^^)~
0 M-«- '
«_> LU
JC
H-r- *
vj— r~" -it.
o o v--
i. CO
4-> •*-> i—
CO C r—
O O T-
0 0 E

JZ
3 J^
CJe 5t
•r- \ JK1
CO "^-^
CO C X
CO O O
na T- z:
cu co
i. 10 10
O •!- .Q
O> E r—
Q UJ- 	

f 	 	
i-
C -C
o •^~
•r— O X
CO +-> O
co to 2:
•i- C£
E CO
UJ JO











cr
o
•r—
4->
«O
O
•f—
"a.
rx
<
0
o
•t*
A
1— 1






r>-
r^
CO
t
CVJ




co
vo
•
«*
«— 4





o
CO
o
o
*









0
•
CVJ


i_
>- o
•^ co
CO
i-
3Z
ai a.
N JC
•r- O
co in
CO

>>
JQ '
•o
tr
(O t— 1
-M 1
co co

•
<
o o
O O
**-  o
CO T-l
*
r-i


"


CM CO
1 1
co co



0
o
«*
ft
r-H






CO
r^
CO
•





CM
VO
•
CVI






o
co
O
o










^
•
vo



o
CD
CM

a.
.c
o
0
.— i
«*
t-i





«±
i
CO




























































flO
•f™*
i-
4J
CO
3
XJ
C
1 — 1

•
CO
     8
o
o
o
            o
            o
            o
o
o
o
      o
      o
      o
§
o
  «%
o
I-.
                         o
                         o
                         o
                                                        CM
     CVJ
            CO
       LO

       Cn

       CVJ
                         o
                         cvi
                         LO
      co
      LO
      CVI
                                           CM
                                                  en
                                                        CVJ
     cvj
     o
            in
                  CVJ
                               o

                               CO
                                     cr>
                                CVJ
                                r-.
                                                         cvj
      o
      o
 o
 o
             o
             o
LO
O
o
cn
in
O
o
              LO
              o
              o
 en
 in
 o
 o
o
o
cvi    cvj    cvi


«*    «=J-    «3-
cvi    cvj    cvi
O    O

vo    vo
cn    cn
in    in
      o
      CD

      O
        A

      CVJ
      O
      O
      o
 o
 o
 o
             CD

             O

             CD
o
o
CVJ
CD
o
in
            CO    00
 CD
 o
 o
              Q.
             JC
             o
             o
             o
VO
10
VO

VO
                                            VO
                                            cn
                                            LO
              o
              CD
              o
               *
              Cvj
              VO

              ID
                                       vo
                                       cn
                                       LO
 o
 o
 o
                                                         CVI
                                                   cvi
o
o
o
                                                   co    co
 vo
 vo
            CVJ

              I
                           CVJ
                            I
                                       co
                                        I
o
o
o
                                                          01


                                                          o
                                                         JC
                                                          I/)
7-86

    -------
7.5  POTENTIAL SOCIO-ECONOMIC AND INFLATIONARY IMPACTS
     The Administrator has determined that for any action resulting in a
greater than 5 percent price increase, an inflation impact statement should
be prepared.  Gas turbines for offshore platforms will cost 7.7 percent more
when equipped with air pollution controls.  Yet a vast majority of the offshore
platforms will utilize turbines smaller than 10,000 hp and will thus be exempt
from the regulation for five years under the recommended standard.  Nevertheless,
1t is likely that some of the larger new platforms will require turbines of
greater than 10,000 hp.  However, these will constitute such a small percentage
of the overall market that the net price effect will be well below five
percent.  Therefore, a formal inflation impact statement is not required.
                                     7-86A

    -------
                               REFERENCES
 1     Adapted from Gas Turbine International  Computer Marketing System
       data base.
 2     Sawyer's Gas Turbine Catalog,  1975, Gas Turbine International.
 3     Carlson, P.G., "Economic Impact of Water Injection for NOX Emission
       Control on Solar Gas Turbines," December 16, 1974.
 4     Farmer, R.C., "Rolls Debuts Industrial  Sprey," Gas Turbine Inter-
       national, May-June, 1974.
 5     Sawyer's Gas Turbine Catalog,  1975, loc. cit.
 6     American Society of Mechanical  Engineers, letter from Paul Hoppe to
       Don Goodwin, Director ESED, EPA/OAQPS,  dated October 9, 1975.
 7     Sawyer's, p. 131
 8     Letter and attachments from Ralph Kress, Solar, to D.R. Goodwin,  EPA,
       dated August 26, 1974.
 9     General Motors response to Preliminary  (draft) Proposed Standards  for
       Control of Air Pollution from Stationary Gas Turbines, March 1973.
10     R.W. Beck and Associates,  "Control of NOX Emissions from a Combustion
       Turbine Plant in South Dakota," prepared for Basin Electric Power    i
       Cooperative, August 1975.                                            !
11     Letter and attachments from Gene Zeltman, General  Electric, to  Kenneth
       Durkee, EPA, October 31, 1975.
12     Letter and Attachments from R.H. Gaylord, Turbodyne to D.R. Goodwin,
       EPA, December 19, 1975.
13     Data submitted by G.E. to  EPA during meeting in September 1975.
14     N.A. Rockwell, letter from Gene Evenson to Jeff Weiler, EEA, dated
       October 10, 1975.
15     Vogel, Robert G., "Analysis of EPA Suggested New Source Performance
       Standards for Stationary Gas Turbines," Southern California Gas  Com-
       pany, March 1975.
16     Carlson, P.G., op.  cit.
17     Culligan, Inc. letter from Al  Lorenzo to Robert Coleman dated March  1976.
18     Ultrascience Corporation,  pamphlet on U.D.I. "7",  Deionizer, April 1975.
                                      7-87

    -------
19    CulUgan, Inc.,  op.  cit.
20    UHrasdence Corporation, op. cit.
21    Carlson.  P u., op.  cit.
22    Culligan, Inc., op. dt.
23    Osmonics, Inc.  letter from  Peter Cartwright to Robert Coleman dated
      December 8, 1975.
24    Arrowhead Water,  letter  from Mr. Laird Lewis to Mr. Jeff Weiler, EEA,
      dated October 31,  1975.
25    Continental Water  Conditioning  Corporation, letter from Robert Taylor
      to Jeff Weiler,  EEA, dated  January 2, 1976.
26    Illinois Water Treatment, letter from Mr. Leonard Snead to Robert
      Coleman, EEA, dated March 1976.
27    Vogel, Robert G.,  op.  cit.
\
28    Ibid.
29    San Diego Gas and  Electric, letter from Jon Hardway to Jeff Weiler,
      EEA, dated October 2,  1975.
30    Arrowhead Water,  op. cit.
31    "Supplement 11  to  Tariff 1009A," Bulk Carrier Conference, Inc., 12001
      Jefferson Davis  Highway, Arlington,  Virginia.
32    Carlson, P.G.,  op.  cit.
33    Continental Water Conditioning Corporation, op. cit.
34    Carlson, P.G.,  op.  cit.
35    Vogel, Robert G.,  op.  cit.
36    Carlson, P.G.,  op.  cit.
37    Gibbs and Hill,  Inc.,  letter from H.F. Sterba to Robert Coleman, EEA,
      dated November 20,  1975.
                                                   i
">8    Iowa Public Service, letter from Mr. Warren Kane to Robert Coleman,  EEA,
      dated March 1976.
39    San Diego Gas and  Electric, letter from Walter Zitlau to Don Goodwin,
      Director, ESED,  EPA/OAQPS,  dated October  19, 1972.
40
      San Dieao Gas and Flectric,  Hardway letter,  op.  cit.
                                      7-!

    -------
41    American Society of Mechanical  Engineers, op. c1t.
42    Carlson, P.G., op. cit.
                                                              •
43    General Motors response, op.  cit.
44    Data submitted by M. Jarvis,  G.E. to EPA November 1972.
45    Vogel,   Robert G., op. cit.
46    San Diego Gas and Electric,  Zitlau letter, op.  cit.
47    R.W. Beck and Associates, op. cit.
48    Gibbs and Hill, op. cit.
49    Simmons Precision, letter from M.D. Nuttrass, Industrial  Product
      Line Manager, to Jeff Weiler, EEA.
50    G.E. Data submitted to EPA during meeting in September 1975.
51    Ibid.
52    Gaylord to Goodwin letter, op.  cit.
53    Letter  and attachments from W.C.  Lee, President of Environics, to
      Kenneth Durfkee, EPA, October 29,  1975.                                  ~~"
                •\L
54    Ibid.                                                          ,
55    General Electric, letter from Gene Zeltman to Eric  Noble,  ESED, EPA/OAQPS
      dated December 17, 1973.
56    Letter  and attachments from R.B.  Snyder, Portland General  Electric,
      to Kenneth Duijlkee, EPA, dated December 26, 1975.
                   i^
57    Letter  and attachments from Donald Dunlop, Vice  President of Florida
      Power Company, to Don Goodwin,  EPA, dated December  1, 1975.
58    "Petroleum Facts and Figures, 1971," American Petroleum Institute.
59    "Refined Product Prices," Oil and Gas Journal,  p. 117, April  12, 1976.
60    "The Costs of Sulfur Oxide Controls to Oil Burning  Power Plants in  1980,"
      report  prepared by EEA for U.S. EPA under contract  68-01-1924, Sept.  1975.
61    "Refined Product Prices," op. cit.
62    "The Costs of Sulfur Oxide Controls to Oil Burning  Power Plants in  1980,"
      op. cit.
                                    . 7-89

    -------
63    "Production of Low Sulfur Gasolines, Task 10 Report, Phase 1," report
      to EPA by M.W. Kellogg Co. under contract No. 68-02-1308, January 1974.
64    "Flue Gas Des"lfurizatiort Process Cost Assessment," report to EPA
      by PEDCo-Environmental under contract 68-01-3150, May 6, 1975.
65    Personal communication from Mr. Chris Lombardy, Teller Corporation,
      to Mr. Robert Coleman, EEA, March 1976.
66    Oil and Gas Journal,  November 3, 1975.
67    .tatement of John Nassikas, FPC before the House Committee on Interstate
      and Foreign Commerce, July 14, 1975.
68    Steam Electric Plant  Statistics, FPC Form 1, 1974.
69    Sawyer, J.W. and Farmer, R.C., "Gas Turbines on Gas Pipelines,"
      Sawyer's Gas Turbine  Catalog, 1975, p.  217
70    Steam Electric Plant  Statistics, op. cit.
71    Continental Water Conditioning Corporation, op. cit.
72    Gibbs and Hill, op.  cit.
73    "Dallas Rate Survey," Journal of the American Hater Works Association,
      May 1975.
74    "Supplement to Tariff 1009A", op. cit.
75    Carlson, P.G., op. cit.
76    Data submitted by General Electric to EPA during meeting in September 1975.
77    Gibbs and Hill, op.  cit.
78    Vogel, Robert G., op. cit.
79    "Preliminary 1974 Generation Figures," Federal  Power Commission News
      Release No. 21450, June 1975.
80    Electric Power Statistics, June 1975, Federal Power Commission.
81    Ibid.
                                      7-90

    -------
                             8.   RATIONALE






8.1  SELECTION OF SOURCE FOR CONTROL



     As described in chapter 3,  stationary gas turbines are sources of parti -



culate, NOV, S09, CO, and HC emissions.  NO  emissions, however, are of more
          X    £                           X


concern than emissions of these  other pollutants for two reasons.   First,



NO  is emitted in greater quantities from stationary gas turbines  than
  A


are these other pollutants.  Second, EPA has assigned a high priority to



development of standards of performance limiting NO  emissions.   Assuming
                                                   X


existing levels of emission controls, national NO  emissions from stationary
                                                 A


sources are projected to increase by about 65 percent by 1985.   Applying



best technology to all new sources would reduce this increase to about



25 percent, but would not prevent it from occurring. This unavoidable



increase in NO  emissions is attributable largely to the fact that
              X


current NO  emission control techniques are essentially design  modifications,
          X


not add-on equipment or operation changes.  In addition, few NO  emission
                                                               A


control techniques can achieve large (i.e., in the range of 90  percent)



reductions in MOV emissions.  Consequently, EPA has assigned a  high
                A


priority to the development of standards of performance for major NO
                                                                    A


emission sources wherever significant reductions in NO  emissions  can be
                                                      X


achieved.



     Several studies sponsored by EPA have classified stationary gas turbines



as major controllable sources of NOY emissions.  One study conducted by the
                                   X


Aerotherm Division of Acurex Corporation in 1975 estimated that oil-fired






                                  8-1

    -------
and gas-fired stationary gas turbines accounted  for  2.5  percent  of the total



NO  emissions from stationary sources in the U.S.  in 1972.   This same study
  X


ranked gas-fired turbines as sixteenth and oil-fired gas turbines as



twenty-third in a priority listing,  by equipment type and firing type, of



1?7 controllable stationary sources  of NO  emissions.
                                         X


     In another study The Research Corporation of New England (TRC) determined



th~ impact which standards of performance would  have on  nationwide emissions



of particulates, NO , S0?, HC, and CO from stationary sources.   Sources
                   X    £-


were ranked according to the impact, in tons/year of pollutant,  which a



standard promulgated in 1975 would have on emissions in  1985.  This



ranking placed gas turbines first on a list of 40 stationary NO  emission
                                                               A


sources.



     In a subsequent study, Argonne National Laboratory  (ANL) expanded the



results of the TRC study to develop a priority listing of sources of



particulates, NO , S09, HC, and CO,  with CO not considered a pollutant
                X    £.


for control by itself.  In developing this listing,  source screening



factors were used which considered items such as:



     •type, cost, and availability of control technology



     •emission measurement methods and applicability



     •enforceability of regulations



     •source location and typical source size



     •energy impact



     •impact on scarce resources



     •other environmental media constraints



     The priority listing, which ranked 237 source-pollutant combinations



in the  order in which standards of performance should be developed, ranked



utility gas turbines fifth and pipeline gas turbines twelfth with NOV being
                                                                    A


the pollutant  identified for control.
                                  8-2

    -------
     In 1974, 90 percent of all domestic stationary gas turbine capacity



was sold to the electric utility market, primarily for use as peaking



units.  It is expected that this large percentage of sales to utilities



will continue in the future due to the many advantages of gas turbines



as peaking units.  Stationary gas turbines are characterized by low capital



cost, ease of installation, minimal maintenance requirements, and low physical



profiles (low buildings and short stacks), all of which make gas turbines



attractive for peaking units.  In addition, turbines can be instrumented for



remote operation, have very low visible emissions, and when properly muffled



are quiet in operation.  As a result, turbine peaking units are often



located in large urban centers where power demands are greatest and pollution



problems are often most severe.



     Stationary gas turbines, therefore, are significant contributors to



total nationwide emissions of NO.  They are ranked high on the various
                                X


listings of sources for which standards of performance should be developed.



In addition, the expanding market for gas turbines coupled with the probability



that many turbines will be installed near large urban centers underscores the



necessity of developing standards of performance for stationary gas turbines.



Consequently, stationary gas turbines are selected for development of standards



of performance.





8.2  SELECTION OF POLLUTANTS



     The pollutants emitted from stationary gas turbines are particulates,



NO , S0?, CO and HC.  The nature and amounts of each of these pollutant
  A    C-


emissions from gas turbines vary a great deal and have been discussed in



chapters 3 and 4.



     As discussed above, stationary gas turbines are a significant source



of NOV emissions, ranking first out of a list of 40 sources of NO
     A                                                           A
                                 8-3

    -------
emissions according to the TRC report and fifth out of 237  source-
pollutant combinations for utility turbine NO  emissions  and twelfth
                                             A
out of 237 source-pollutant combinations for pipeline turbine NO  emissions
                                                                A
according to the ANL report.   Combustor modifications (dry  control) and
water injection (wet control), as discussed in chapters 4 and 7, are
demonstrated techniques for reducing NO  emissions at reasonable cost
                                       A
and, as discussed in chapter 6, depending on the specific emission level
selected, could reduce NO  emissions by as much as 190,000  tons per year
                         A
in 1982.  This is a significant decrease in total  nationwide NO  emissions.
                                                               /\
For these reasons, NO  emissions from stationary gas turbines are selected
                     A
for control by standards of performance.
     SOp emissions from stationary gas turbines depend on the sulfur content
of the fuel since nearly 100 percent of the sulfur in the fuel is converted
to SOp during combustion.  The TRC report ranked gas turbines 8 out of
41 major contributors of SOp emissions.  Due to the high volumes of
exhaust gases, as discussed in chapter 7, the cost of flue  gas desulfurization
(FGD) to control SOp emissions from stationary gas turbines is considered
unreasonable.  Selection of low sulfur fuels, however, is considered
reasonable.  Control of SOp emissions, therefore, requires  combustion of
low sulfur fuels rather than the application of FGD.
     Since stationary gas turbines are a major source of SOp emissions  and
selection of low sulfur fuels is considered an economically feasible
control technique, SOp emissions from stationary gas turbines are selected
for control by standards of performance.
                               8-4

    -------
     Hydrocarbon and carbon monoxide emissions from stationary gas turbines
operating at full load are relatively low because the higher the percentage
of full load at which a turbine operates, the more efficient the combustion
of the fuel and the lower the emissions of HC and CO.  Gas turbines
normally operate at 80 to 100 percent of full load with HC emissions
averaging less than 20 ppm and CO emissions averaging less than 150 ppm
at ,15 percent Op.  On occasion utility gas turbines may operate at 20-30
percent of load when in the "spinning reserve" mode.  As shown by the
data presented in Appendix C, at 10-20 percent of full load, HC emissions
from large utility gas turbines are usually less than 50 ppm and CO
                                                                    on
emissions are usually less than 500 ppm at 15 percent 0?.  Informati
submitted to EPA by the Edison Electric Institute,  however, indicates
that utility gas turbines very seldom operate in the "spinning reserve"
mode due to economic considerations.  This conclusion was also confirmed
in telephone conversations between utility and EPA personnel.
     HC emissions from gas turbines were not included in the rankings of either
the TRC or ANL reports.  CO emissions from gas turbines were ranked sixth
out of 42 contributors of CO in the TRC report but were not ranked in
the more comprehensive ANL report.
     Gas turbines will be operated as efficiently as possible to conserve
fuel which at the same time will reduce HC and CO emissions; therefore,
HC and CO emissions from stationary gas turbines are not selected for
control by standards of performance.
                              8-5

    -------
     Participate emissions from stationary gas turbines are minimal as



shown by the TRC report which ranks particulate emissions from gas



turbines 111 out >f 113 particulate emitters, and the ANL report which



ranks utility and pipeline gas turbines 228 and 229, respectively, out



of 237 possible source-pollutant combinations for particulate emissions.



Consequently, particulate emissions from stationary gas turbines are not



se,?cted for control by standards of performance.





8.3  SELECTION OF AFFECTED FACILITIES



     Stationary gas turbines are produced in three different configurations:



simple cycle, regenerative cycle, and combined cycle.  All of these configura-



tions emit NO  and S0~, and all can be controlled for NO  emissions by
             A       C.                                  X


water injection or dry controls and for S0? by firing low sulfur fuels.



Consequently, simple cycle turbines, regenerative cycle turbines and the



gas turbine portion of combined cycle plants are selected as affected



facilities for standards of performance limiting NO  and S0? emissions.
                                                   A       C.


     Gas turbines can burn either liquid or gaseous fuels.  The dry and



wet control techniques described in chapter 4 for the control of NO
                                                                   /\


emissions can be applied to gas turbines regardless of the type of fuel



burned.  Similarly, the firing of low sulfur fuel for the control  of



S02 emissions can be applied to gas turbines regardless of the type of



fuel burned.  Consequently, gas turbines burning all types of fuels are



selected as affected facilities for standards of performance limiting



NO  and S09 emissions.
  X       £


     As discussed earlier, 90  percent of all stationary gas turbine capacity



was sold to the electric utility industry in  1974 and this trend  is expected



to continue.  In most cases these gas turbines were large units of 10,000  hp




                                 8-6

    -------
or more.  Most of the remaining 10 percent of the gas turbine capacity sold



in 1974 consisted of units in the range of 1000-10,000 hp.  Thus, the con-



tribution of small gas turbines of less than 1000 hp to national NO  emissions
                                                                   rt


is negligible.  For many applications up to about 10,000 hp, stationary



gas turbines compete with stationary internal combustion engines.  A



standard of performance on one of these industries and not the other,



therefore, would tend to give the non-regulated industry a competitive



advantage to some extent.



     Currently, standards of performance are being developed for stationary



internal combustion engines.  Although relatively few internal combustion



engines of greater than 1000 hp are produced, these engines are responsible



for 75 percent of the total NO  emissions from stationary internal combustion
                              /\


engines.  Under 1000 hp, however, the number of internal combustion engines



produced increases tremendously and enforcement of standards of performance



would not be feasible in the absence of a certification program similar to



that for automobiles.  Since the Clean Air Act does not permit standards



of performance to be enforced by a certification program, a lower size cutoff



of 1000 hp for standards of performance for stationary internal combustion



engines is considered appropriate.  Consequently, to be consistent a



lower size cutoff of 10.7 gigajoules per hour heat input  (about 1000 hp)



is selected for standards of performance for stationary gas turbines.



Below this cutoff the standards limiting NO  and S0? emissions would not
                                           A       Cm


apply.



     Some gas turbines are operated as a mechanical or electrical power



source only when the primary power source for a facility has been rendered



inoperable by an emergency situation.  This type of turbine operates



infrequently, usually only for checkout and maintenance; therefore, it




                                  8-7

    -------
contributes only a very small  amount to total  nationwide NO  emissions
                                                           A


emitted by all gas turbines.   Since this type  of turbine is operated infrequently,



the owner of sue*- a turbine would probably choose,  for economic reasons, to



store water for injection rather than installing a  water treatment unit on



s te.  If the emergency situation was of such  duration that the turbine



used the entire supply of injection water before it could be replenished, then



th.. turbine would have to shut down, leaving the facility the turbine



served with no source of power.  This situation could possibly be dangerous



in some cases, such as where turbines are used to supply emergency power



to hospitals.  There also could be operational problems with the water



injection system due to the long periods of non-operation.  Consequently,



emergency standby gas turbines are exempted from standards of performance



limiting NO  emissions.
           /\


     Gas turbines could possibly contribute to the creation of ice fog,



which consists of small (mean diameter 3 to 7 microns) ice crystals which



are nucleated by airborne particulate.  Ice fog occurs at temperatures  below



-28°C and is a serious problem in only a small portion of the United  States,



primarily Alaska.  Ice fog severely restricts visibility and, since the



crystals are long-lived, can plague auto and air traffic for extended periods.



The actual impact of water or steam injection on the formation of ice fog



is unknown; however, water or steam injection will increase the moisture



content of the exhaust gas discharged by gas turbines.  Since ice fog



occurs only in a small portion of the United States and only under



special weather conditions, the  impact on air quality due to increased



NO  caused by exempting gas turbines  creating ice fog would be minimal.
  /\


Therefore, gas turbines using water or steam injection for  control of



NO  emissions are exempted from  the standards limiting NO   emissions when
  A                                                      X


ice fog created by the turbine is deemed by the owner or operator of  the



turbine to be a traffic hazard.




                                   8-8

    -------
     Stationary gas turbines are sometimes used by the military in combat-



type situations.   The main advantage of these turbines is their mobility,



which would be considerably restricted by a water injection system consisting



of either water treatment equipment or a water storage vessel.   Restriction



of the mobility of these turbines could have an adverse effect  on national



defense; therefore, any military combat-type gas turbine for use in other



than a garrison facility is exempt from the standards limiting  NO  emissions.
                                                                 A


     The possibility of exempting some gas turbines from the standard limiting



SO^ emissions was also examined.  Except for exempting all turbines of less



than 10.7 gigajoules per hour heat input (about 1000 hp), no exemptions



from the S0? emission limit were considered necessary.





8.4  SELECTION OF THE BEST SYSTEM OF EMISSION REDUCTION



     As discussed in chapter 4, there are three possible control techniques



for reducing NO  emissions from stationary gas turbines:  wet controls, dry
               X


controls, and catalytic exhaust gas cleanup.  Wet controls involve the



injection of water or steam into the combustion reaction to reduce peak



flame temperatures, thereby reducing NO  formation.  Wet control techniques
                                       /\


have been demonstrated on many large gas turbines (greater than 10,000



hp) used in utility and industrial applications, and these installations



have had good reliability over long periods of operation.  Wet controls,



however, have not been demonstrated on small production gas turbines



(less than 10,000 hp), although the effectiveness of these techniques



for small gas turbines has been evaluated in laboratory and combustor



rig tests.  Thus, wet controls can be applied immediately to large



stationary gas turbines, but manufacturers estimate that at least three



years are required to incorporate and demonstrate wet control techniques



on small production gas turbines.





                                   8-9

    -------
     Dry controls consist of operational  or design modifications which



govern combustion conditions to reduce NO  formation.   Although dry
                                         A


controls have been tested in laboratory and combustor rig tests, manufacturers



estimate that up to five years is required for further development,



design, test, and incorporation of dry controls on large and small



stationary gas turbines.  Catalytic exhaust gas cleanup consists of NO
                                                                      A


reduction by ammonia in the presence of a catalyst.  While laboratory



tests are very promising, this technique is not demonstrated.  Consequently,



only wet controls and dry controls are considered as viable alternatives



for development of standards of performance.



     The NO  emission reduction achievable with these two alternatives clearly
           A


favors the development of standards of performance based on wet controls.



Reductions in NO  emissions of more than 70 percent have been demonstrated
                /\


using wet controls.  Dry controls, however, have demonstrated NO  emission
                                                                J\


reductions of only about 30 percent.  Laboratory tests have indicated that



some dry control techniques have the potential of achieving much greater



NO  emission reductions, but these techniques need considerable development
  X


before they can be considered demonstrated.



     Standards of performance based on wet controls would reduce national



NO  emissions by about  190,000 tons per year in 1982.  In contrast, standards
  A


of performance based on dry controls would have no impact on national NO
                                                                        X


emissions in 1982, due  to the necessity of allowing a five-year delay to



incorporate dry controls on gas turbines.  By 1987, standards based on



wet controls would reduce national NO  emissions  by about 400,000  tons per
                                     X


year, whereas standards based on dry controls would reduce NO  emissions by
                                                             /\


only about 90,000 tons  per year.  Thus, standards of performance based



on wet controls would have a much greater  impact  on national NO  emissions
                                                               A


than standards based on dry controls.






                                  8-10

    -------
     Uncontrolled ambient NO  levels near stationary gas turbines are typically
                            A
                                                                3

well below the National Ambient Air Quality Standard of 100 yg/m  due primarily



to the high temperature of the exhaust gases and resulting rapid plume rise.



Typically, uncontrolled annual average ambient NO  concentration levels
                                                 X


from dispersion model studies of stationary gas turbines range from below 1 to

             3

about 14 yg/m .  One model plant calculation, however, yielded an uncontrolled

                                  3

ambient NO  level of about 50 yg/m .  Standards of performance based on
          A

                                                                     3

wet controls would reduce this ambient concentration to about 10 yg/m



while standards based on dry controls would only reduce this ambient NO
                                                                       A
                                    3

concentration level to about 30 yg/m .  Where ambient NO  concentration
                                                        A


levels near stationary gas turbines would be significant, therefore,



standards of performance based on wet controls would be more effective



in reducing ambient NO  concentration levels than standards of performance
                      A


based on dry controls.



     The water pollution impact of standards based on wet controls would



be minimal.  Water needed for wet controls may be treated by the same



processes used to treat steam boiler make-up water.  The quality of the



wastewater from this treatment is essentially the same as the influent



water except that the concentration of total dissolved solids in the



effluent stream is 3 to 4 times that of the influent.  In most cases, the



effluent may be sewered directly or returned to the river supplying the



water.  Where this is not possible, the effluent may be discharged to



an evaporation pond.  Consequently, the water pollution impact of standards



based on wet controls would be minimal.



     The quantity of water required by a stationary gas turbine using



wet controls is relatively small.  The upper limit water-to-fuel ratio



of about 1:1 requires only about 5 percent of the quantity of water





                                  8-11

    -------
consumed by a comparable size steam boiler using  cooling  towers.   A
water treatment system for five 28 MW stationary  gas  turbines  operating
10 hours per day using a water-to-fuel  ratio of 1:1,  for  example, would
treat 125,000 gallons of water and reject about 25,000 gallons of wastewater
per day.  A steam boiler of comparable  size with  cooling  towers would
consume about 20 times as much water.  In fact, the usage rate of water
fo  wet controls is small enough that the unlikely prospect of having to
truck water 50 miles was determined to be economically reasonable as
discussed below.  Standards based on dry controls, however, would have
no impact on water pollution or water supplies.
     Standards based on wet controls would have a negligible solid waste
impact.  When it is not possible to sewer the wastewater effluent directly
or to return it to the source from which the water was drawn,  the use of
an evaporation pond will result in the slow buildup of solids  in the pond.
These solids can be periodically collected and disposed of in  landfills.
Standards based on dry controls, however, would have no solid waste impact.
     There would be no adverse noise impact resulting from standards based
on either wet or dry controls.
     The potential energy impact of standards based on wet controls is small.
Standards based on wet controls could increase the fuel consumption of a
gas turbine by as much as 5 percent, depending on the rate of water injection
required to comply with the standard.  Few turbines will require the high
water injection rates  (about 1:1 water-to-fuel ratios) which result in
a 5 percent fuel penalty.  Assuming that all stationary gas turbines
subject to compliance with standards would require a 1:1 water-to-fuel
ratio,  the fifth-year energy impact on large stationary gas turbines would
be about 5500 barrels of fuel oil per day in 1982.  The fifth-year energy
                                   8-12

    -------
impact on small stationary gas turbines would be about 7000 barrels per
day in 1987, as a result of the delayed effective date of the proposed
standards on small turbines.  Each increase represents less than a 0.03
percent increase in estimated oil consumption in the United States in
1982 and 1987.   It should also be recognized that these energy impacts
are based on assumptions which yield the greatest energy impacts.
Actual energy impacts are expected to be much lower.  The energy impact
of standards based on wet controls, therefore, is minimal.  Standards
based on dry controls, however, would have no energy impact.
     Although wet controls do result in a small adverse impact on gas turbine
efficiency, the costs associated with this increased fuel consumption may,
for some applications, be partially offset by an increase in the gas
turbine's rated power output capability.  Based on manufacturer's estimates,
gas turbine baseload capacity will be increased by 3 to 4 percent as a
result of water injection.  In applications where turbines are operated
at maximum capacity, such as utility power generation and pipeline compressor
stations, this increased baseload capacity essentially reduces the installed
costs per kilowatt by the percentage increase in the capacity of the unit,
thus slightly reducing the cost impact of standards based on wet controls.
     The economic impacts associated with standards based on either wet
or dry controls would be small.  Dry control costs are difficult to quantify.
Many manufacturers, however, have indicated that the cost of dry controls
would not exceed the cost of wet controls.  Consequently, the analysis
of the economic impact of standards of performance is based on the costs
of wet controls and assumes that the costs of dry controls, and hence the
economic impact of standards based on dry controls, would be comparable.
Standards of performance, therefore, based on either wet or dry controls
                                  8-13

    -------
would increase the capital  cost of a gas turbine for most applications by
about 1 to 4 percent.   For  offshore industrial  applications where desalini-
zation equipment is required to provide water for wet controls, standards
would result ii  a 7 percent increase in the capital  cost of a gas turbine.
Annualized costs for all applications would be increased from 1 to 4
percent, with utility applications realizing less than a 2 percent
increase.
     Although it is unlikely that a stationary gas turbine would, of
necessity, be installed in  an arid area, an analysis was performed which
assumed that water would have to be transported to the gas turbine site
by truck over a distance of 50 miles.  This unlikely situation would result
in less than a 4 percent increase in the annualized cost of the gas turbine.
     Standards of performance based on wet controls would increase the total
capital investment requirements for all industrial and commercial users of
large stationary gas turbines (greater than 10,000 hp) by about 36
million dollars by 1982.  Total annualized costs would be increased by
about 11 million dollars per year in 1982.  Standards of performance
based on wet controls would have an additional economic impact on users
of small stationary gas turbines  (less than 10,000 hp) beginning in
1982.  Thus, for the period of 1982 through 1987, the capital  investment
requirements for all stationary gas turbine users would be about 67
million dollars.  The annualized costs would be about 30 million dollars
by 1987.  These impacts would translate into price increases for the  end
products or services provided by these  industrial and commercial users
of stationary gas turbines ranging  from less than 0.01 percent in the
petroleum refining industry to about 0.1 percent  in  the electric utility
industry.  Thus, the economic impact of standards of performance based
on wet controls would be very small.

                                   8-14

    -------
     Standards of performance based on dry controls would have no economic
impact on users by 1982.  Following 1982, however, the economic impact
of standards based on dry controls would be comparable to that of standards
based on wet controls.
     Based on this assessment of the impacts of standards of performance
based on wet controls and dry controls, wet controls are selected as the
"...best system of emission reduction (considering cost}..." for the reduction
of NO.
     A
     As discussed in chapter 4, there are two possible control techniques
for reducing SOp emissions from stationary gas turbines:  flue gas desulfurization
(FGD) and the firing of low sulfur fuels.  FGD, however, as pointed out
in chapter 7, would cost about two to three times as much as the gas turbine.
The economic impact of standards of performance based on FGD, therefore,
is not considered reasonable.
     Low sulfur fuels, such as premium distillate oils or natural gas,
are now being burned by nearly all stationary gas turbines.  These premium
fuels are being burned primarily because the increased maintenance costs
associated with firing heavy fuel oils are greater than the savings that
would be realized by buying these less expensive heavy or residual fuel oils.
Over the next five to ten years, however, as oil prices continue to
escalate, the price differential between premium distillate fuel oils
and heavy fuel oils will increase.  The economic incentive to burn the
premium fuels, therefore, will probably become marginal.
     In the absence of regulations requiring gas turbines to fire specific
fuels, the choice between firing either premium distillate fuel oils or
heavy fuel oils will  likely be decided on the basis of the relative
convenience and availability of these fuels.  Premium distillate fuel oils
                                   8-15

    -------
are more convenient to burn than heavy fuel  oils  because they have a
lower viscosity and are easier to handle.   Heavy  fuel  oils frequently
require heating, for example, to reduce their viscosity to the point
where they can be readily pumped from one  location to  another.  Even if
the price differential between premium distillate fuel oils and heavy fuel
oils were to increase to the point where the firing of heavy fuel oils was
marqinally attractive, the greater inconvenience  of scheduling and
performing the additional maintenance would probably cause a gas turbine
user to choose to fire the premium distillate fuel oil.  On the basis of
convenience, therefore, stationary gas turbines are likely to continue
firing premium distillate fuel oils even if the economic incentive to do
so becomes marginal.
     The impact on ambient air quality of standards of performance based
on the firing of low sulfur premium distillate fuel oils in gas turbines,
therefore, would be negligible.  The economic impact would also be
negligible for the same reason and there would be no water, energy,
solid waste or noise impact associated with standards  based on the
firing of low sulfur premium distillate fuel oils.
     Based on this assessment of the impacts of standards of  performance
based on the firing of low sulfur fuel oils, this control technique  is
selected as "...the best system of emission reduction  (considering
costs)..." for the reduction of S02 emissions.

8.5  SELECTION OF FORMAT FOR THE STANDARDS
     A number of different formats could be selected  to  limit NO   emissions
                                                                 /\
from stationary gas turbines.  Mass standards limiting emissions  in  terms
of power output  (i.e. mass of emissions per unit  of power output)  or con-
centration standards  limiting the concentration of emissions  in  the  exhaust
gases discharged into the  atmosphere could be developed.
                                   8-16

    -------
     While mass standards may appear more meaningful  in the sense that they
relate directly to the quantity of emissions discharged into the atmosphere,
enforcement of mass standards is more costly and the results more subject
to error than enforcement of concentration standards.  Determining mass
emissions, for example, requires measurement of power output and exhaust
gas flow rates in addition to the measurements required for a concentration
standard.  Power output can be readily obtained at most electric utility
installations, but shaft power at compressor and industrial installations
would be difficult and expensive to measure accurately.  Also, the high
turbulence characteristic of gas turbine exhaust gases makes the
determination of exhaust gas flow rates, and hence mass emissions, subject
to considerable error.  Manipulation of this data increases the number
of calculations necessary, compounding the errors inherent within the
data and increasing the chance for human error.
     Enforcement of concentration standards, however, requires a minimum
of data and calculations, decreasing the costs and minimizing the chances
for error in determining compliance.
     The primary disadvantage normally associated with concentration standards
is that of possible circumvention by dilution of the exhaust gases discharged
to the atmosphere lowering the concentration of emissions, but not reducing
the total mass emitted.  Thus, concentration standards must be written to
insure that the standards are not met merely by addition of dilution air.
For combustion processes, this can be accomplished by correcting measured
concentrations to a reference concentration of 0?.  The 0? concentration
in the exhaust gases is related to the excess  (or dilution) air.  Typical
Op concentrations in gas turbine exhaust gases are about 15 percent.  Thus,
                                  8-17

    -------
referencing the standard to 15 percent oxygen  effectively  precludes



circumvention by dilution.   Consequently,  a  concentration  standard referenced



to 15 percent oxgen is selected as  the format  for  standards  of performance



for stationary gas turbines.



     Selection of a concentration format,  however, could penalize high



efficiency gas turbines.  Higher efficiencies  are  normally achieved  by



increasing combustor operating pressures  and temperatures  and NO
                                                                X


formation generally increases exponentially  with increased pressure  and



temperature.  High efficiency turbines, therefore, generally discharge



gases with higher NO  concentrations than  low efficiency turbines,.   A
                    A


concentration standard based on low efficiency turbines could restrict



the use of some high efficiency turbines.   Conversely, a concentration



standard based on high efficiency turbines could allow such high NO
                                                                   A


concentrations that low efficiency turbines  would  require  no controls.



Consequently, having selected a concentration format for standards of



performance, an efficiency adjustment factor needs to be selected to



permit higher NO  emissions from high efficiency gas turbines.
                A


     As mentioned above, NO  emissions tend  to increase exponentially
                           A


with increased efficiency.   It is not reasonable from an emission control



viewpoint, however, to select an exponential efficiency adjustment



factor.  Such an adjustment would at some point allow very large increases



in emissions for very small increases in efficiency.  The  objective of an



efficiency adjustment factor should be to give an emissions credit for



the lower fuel consumption of high efficiency gas  turbines.  Since the



relative fuel consumption of gas turbines varies linearly  with efficiency,



a linear efficiency adjustment factor is selected to permit increased



NO  emissions from high efficiency gas turbines.  A linear efficiency
  A
                                   8-18

    -------
adjustment factor also effectively limits NO  emissions to a constant
                                            A
mass emission rate per unit of power output.
     The efficiency adjustment factor needs to be referenced to a baseline
efficiency.  Since most existing simple cycle gas turbines fall in the range
of 20 to 30 percent efficiency, 25 percent is selected as the baseline
efficiency.  The efficiency of stationary gas turbines is usually
expressed in terms of heat rate which is the ratio of heat input, based
on lower heating value (LHV) of the fuel, to the mechanical power output.
The heat rate of a gas turbine operating at 25 percent efficiency is
14.4 kilojoules per watt-hr (10,180 Btu per hp-hr).  Thus, the following
linear adjustment factor is selected to permit increased NO  emissions
                                                           A
from high efficiency stationary gas turbines:
                             v  - v 14'4
                             xa " x ~T~
     where:
     x, = adjusted N0v emissions permitted at 15 percent oxygen and ISO
      a              x
          conditions, ppmv.
     x  = NO  emission limit specified in the standards at 15 percent
            A
          oxygen and ISO conditions, (i.e. 75 ppmv).
     Y  = LHV heat input per unit of power output  (kilojoules/watfhr).
     NOTE:  ISO conditions refers to standard atmospheric conditions of
            760 mm mercury, 288° Kelvin and 60 percent relative humidity.
     The only intent of this efficiency adjustment factor is to permit a
linear increase in NO  emissions with increased efficiencies above 25
                     /\
percent.  Consequently, the adjustment factor would not be used to
adjust the emission limit downward for gas turbines with efficiencies of
less than 25 percent.
                                 8-19

    -------
     The selection of this efficiency adjustment factor will  essentially



preclude the development of gas turbines which achieve higher operating



efficiencies at the expense of exponential increases in NO  emissions.
                                                          X


As a result, th'io linear adjustment factor will require the development



of effective NO  controls on future high efficiency gas turbines.
               A


     The rationale for the selection of a format for the limiting of SO
                                                                       /\


emissions from gas turbines is much the same as that discussed above for



NO  emissions.  Thus, to be consistent with the format selected for standards
  )\


limiting NO  emissions, a concentration standard is chosen as the format
           /\


for the SOp standard.  An emission limit in terms of percent fuel sulfur



content has also been included in the S02 standard to give the owner or



operator the flexibility of either measuring the S02 concentration of the



exhaust gas or analyzing the fuel being fired in the turbine.  Either



format for the S02 standard can be used since nearly all of the sulfur



in the fuel is converted to SOp.



     The efficiency factor associated with the NO  emission limit,
                                                 /\


however, will not apply to the SOp emission limit because S02 emissions



do not vary with turbine efficiency.





8.6  SELECTION OF EMISSION LIMIT



     Selection of the NO  emission limit is based on the data and information
                        /\


discussed in chapter 4.  A detailed tabulation of the data may be found  in



appendix C, Figure 4-13  (page 4-25), which is a summary of the wet control



data reproduced here as Figure 8-1.



     While all of these data do not represent maximum NO  reduction efforts,
                                                        /\


the data do indicate the general range of controlled NO  emissions.   Con-
                                                       /\


sidering only the data which represent major NO  control efforts  (i.e.,
                                               A




                                   8-20

    -------
200


180


160


140








—
















O315
11 " T ' .---..

s.
9 ^



9
9
9
rt-

£•120
+->
C
QJ
u
O)
ex
in
-<
t 100
E
o.
o.
co"
o
V)
1 80
UJ
x
o
3E


60

40

20

o
_t_



H —





f
C
1






\
1
f











,0.5
I "














_ <







>





<
,0.6
O
10.9
(j




|)








052
>! -f















r» r «*
i'




N
f








0 6
C






)

9



















0.7
O





r
\




















LEGEND
^ [ ] Combustor rig test




	 Amount of reduction
0.5, etc. Water/fuel
NOTES '
ratio

O The lack of brackets on a
G.T. Size notation indicates
a field or engine test.



9


0.2
9 ?




0.5
' 9

IDS
. J.
6" 9

1.0
y




0.7
9 J°-75
f t <
1 1
!i-i M o t
T 9


9







O-5 |0.5
9
0.75
i"




FACILITY CODE G2 P VI V2 V3 W Y 22 FA HA1 U1 U2 VI
TABLE NO. 8 21 28 30 31 33 35 37 45 47 25 26 28
G.T. SIZE 0.5 2
FUEL TYPE 	 	
.5 17.2 17.2 17.5 21.3 33 33 52 60.4 13 13 17.2

___





O
— j-

9
-4


0.31
(*
-j-
i
1
I
p 0
1
°i
1.01
C

V4 HA2
32 48
17.5 61.5
LIQUID FUEL (DISTILLATE) 	 -». — 	 NATURAL GAS . — *J
—•	—'--	   •••' •   •• t.iuuiu run- vuio i ILU« i t/—	—   •••• 	•• •-    m* *m	ivMiunMU ur\o	
Figure  8-1. Summary of NOX emission data from gas turbines using wet control techniques.
                                       8-21

    -------
water-to-fuel ratios of about 0.6 or greater),  the controlled NO  emissions
                                                                /\


range from about 15 to 60 ppmv.   Much of the variation in these controlled



NO  emission levels can be attributed to such factors as variation in
  A


combustor geome ries, fuel injection systems, water injection techniques,



compression ratios, and combustor inlet air temperatures.



     The available data on emissions from gas turbines using wet controls



come primarily from simple cycle gas turbines and combustor rig tests.  No



reliable data is available concerning NO  emissions from regenerative cycle
                                        X


gas turbines using wet controls, although some dry control data was obtained.



Careful consideration, therefore, must be given to the question of whether



regenerative cycle gas turbines can be controlled to the same emission



levels as simple cycle gas turbines.



     There is general agreement that wet controls will give essentially



the same percentage reduction in NO  emissions from regenerative cycle
                                   A


gas turbines as from simple cycle gas turbines.  Thus, the question



becomes whether uncontrolled NO  emissions from regenerative cycle gas
                               A


turbines are higher than those from simple cycle gas turbines.  On



first comparison, NO  emissions from regenerative cycle gas turbines
                    X


appear higher than those from simple cycle gas turbines.  Regenerative



cycle gas turbines, however, frequently operate at higher thermal efficiencies



than simple cycle gas turbines, and when NO  emissions are plotted
                                           X


against gas turbine thermal efficiency as in Figure 8-2, emissions from



regenerative and simple cycle gas turbines do not appear significantly



different.  As a result, the application of wet controls to either



regenerative or simple cycle gas turbines of comparable  thermal efficiencies



should reduce NO  emissions to essentially the same  level.  Consequently,
                A


regenerative cycle gas turbines should be subject to the same  emission



limit as simple cycle gas turbines.
                                  8-22

    -------
till

o o •» •


0

^ ••
A^A A A
•• • •
o
0 0
O 0* 0 •

O OOCOQO O 


oo
Q3000CDOOO
"~ eo

to
O


+J (O
(C 0)
0 1—

(1J 01
""" *o £
o ea
O) 0)
•r- O
rO O
S-
 E
QC CO
1 I
• O


O O 

O


O OD OO


" •






••
0




1 1 1 1







in
CO



•4-»
0 C
CO O)
o
o>
o.
««
>-
<_J
z
LU
in i— i
C\J CJ
»— 1
ll
1 .
U-
UJ
_J
2»
LU
O 3C
CM »-






in
!"••





0000
O O O O
**• CO CSJ r-
to
C
O
w
"i
0)
X
0
z

a)
i
C
o
u
C
33

C
ID
O
C
o
•?"-


0)
O)
%
3:
01
X)
Q.
"S
C
o
•f—
4-*
IO
*iS
oc

cvj
ci
1

T-
li

Auidd 'SIOdiNOD 13M
                     N39AXO
iN3Dy3d SI e) SNOISSIW3   ON

    -------
     The data also indicate that gas turbines  firing gaseous  fuels typically



have slightly lower controlled NO  emission levels than gas turbines firing
                                 A


distillate fuels.   Again, considering only the data representing major



NO  control efforts, controlled emissions from gas turbines firing gaseous
  A


fuels range from about 15 to 50 ppmv, while controlled emissions from



gas turbines firing distillate fuels range from about 25 to 60 pprnv.



This slight difference in controlled emission  levels does not warrant



the selection of a separate emission limit for each type of fuel.  Only one



emission limit, therefore, will be selected which applies to gas turbines



firing natural gas or premium distillate fuel  oils.



     Based on the emission data and the above  considerations, and allowing for



some uncertainty in the limited data base, 75  ppmv NO  corrected to 15
                                                     /\


percent oxygen is selected as the numerical emission limit for stationary



gas turbines.



     The gaseous and premium distillate fuels  which have traditionally



been burned in stationary gas turbines contain little or no "fuel-bound"



or "organic" nitrogen.  Total NO  emissions from any combustion  source
                                A


including stationary gas turbines, however, are a function of both



thermal NO  and organic NO  formation.  Thermal NO  is formed in "a  well
          XX                       X


defined high temperature reaction between nitrogen and oxygen from  the



combustion air.  Organic NO  , however, is formed by the combination of
                           A


fuel-bound nitrogen with oxygen during combustion.  The reaction mechanism



is not fully understood.  Wet controls are effective for reducing thermal



NO , but are not effective for reducing organic NO  .
  A                                               X


     The emission data presented in Figure 8-1 come primarily from  tests  on



stationary gas turbines firing traditional premium gaseous or distillate



fuels which contain little or no fuel-bound nitrogen.  The measured NO
                                                                      X


                                  8-24

    -------
emissions, therefore, are comprised primarily of thermal NO .   Thus, the
                                                           A


numerical emission limit selected above is not appropriate for the firing



of fuels containing significant amounts of fuel-bound nitrogen.



     Figure 8-3 illustrates the variation in the fuel-bound nitrogen content



of various petroleum fuels.  Although this figure is based on limited data,



it shows the large difference between the fuel-bound nitrogen levels of the



premium distillate fuels and various heavy petroleum fuels.  Generally speaking,



heavy petroleum fuels are not readily available with low fuel-bound nitrogen



levels.  Crude oils generally range between 0.1 and 0.2 percent nitrogen.



As a point of reference, this figure indicates that about half of all heavy



petroleum fuels contain less than 0.25 percent nitrogen.



     Quantifying the organic NO  contribution to total gas turbine NO  emissions
                               A                                     X


is complicated by the fact that the percentage of fuel-bound nitrogen converted



to organic NO  varies with the fuel-bound nitrogen level.  Figure 3-37 (page 3-88),
             A


reproduced here as Figure 8-4, illustrates the variation in conversion of



fuel-bound nitrogen to organic NO  with fuel-bound nitrogen level of the fuel.
                                 A


While this figure is also based on very limited data, it indicates that the



percentage of fuel-bound nitrogen converted to organic NO  decreases as fuel-
                                                         A


bound nitrogen level increases.  Below a fuel-bound nitrogen level of about



0.05 percent, essentially 100 percent is converted to NO .  Above a fuel-bound
                                                        A


nitrogen level of about 0.4 percent, only about 40 percent is converted to NO .
                                                                             A


Using Figure 8-4, an estimate of the effect on controlled NO  emission levels
                                                            A


of firing fuels with various fuel-bound nitrogen levels in gas turbines is



presented in Figure 8-5.  This figure illustrates both the organic NO  and
                                                                     A


thermal NO  contributions to total NO  emissions, assuming that thermal
          A                          A


NO  is controlled to 75 ppmv through the use of wet controls.  As fuel-
  A


bound nitrogen level increases, the organic NO  contribution becomes increasingly
                                              A
                                   8-25

    -------
                                                                                        3
                                                                                        it-
                                                                                         s.
                                                                                         C
                                                                                         S

                                                                                    Q_  CD
                                                                                         e
                                                                                         o
                                                                                         _a
                                                                                         O)
                                                                                         o
                                                                                         4J
                                                                                         CO
                                                                                          I
                                                                                         00
CD
                                                                 CD
                                   8-26

    -------
                                                                      oo
  X
o
                                                                                    x
                                                                                   o
                                                                      C\J
                                                                             >-
                                                         CD

                                                         o

                                                         o
ce:
                                                                             UJ
                                                                             CJ
                                                                                    (D
                                                                                    cn
                                                                                    o
CJ3
O
                                                                      oo
                                                                      CD
                                                                             UJ
                                                                             a.
                                                 LLJ
                                                 cs
                                                 O
                                                 Qi
                                                                      UD    ~
ID

C
2

O
X!
 I


OJ
                                                                            CD
                     o
                     CO
     o
                 CD
                             CD
                             CNJ
01
                                             N390M1IN
                                                                     CNJ
                                                                     CD
                                                                    • CD
                                                                                   to

                                                                                   OJ

                                                                                   C
                                                                                   o
                                                                                   u
                                                                                   O

                                                                                   •£
                                                                                   it!

                                                                                   si
                                                                                   co
                                                                                   g!
                                                                                   CD


                                                                                   LL
                                 8-27

    -------
                                                     OJ
                                                     >

                                                     0!
                                                     G.
                                                     OJ
                                                     O
                                                     TD

                                                     C
                                                     3

                                                     O
                                                i—.   3

                                                LU   «•-
                                                 t   £

                                                 £   £
                                                 O


                                                 £   .e
                                                 D-   w
                                                UJ
                                                     O)
                                                      x.
                                                     o
                                                     -a
                                                     ai
                                                     e
                                                «-^>  O
                                                PQ  o
                                                y  «
                                                U_  co

                                                     cu

                                                     3
                                                     cn
            'ON
8-28

    -------
significant to the point where organic NO  emissions are greater than controlled
                                         X


thermal NO  emissions.
          A


     Three alternative approaches emerge to address the fuel-bound nitrogen



contribution to total NO  emissions from gas turbines.  The first alternative
                        A


is to exempt heavy fuels from standards of performance.  This approach



would allow gas turbines firing heavy fuels to operate with no emission



controls.  In addition to the difficulties of distinguishing between premium



and heavy fuels in the standards, this approach would tend to encourage users



to burn heavy fuels as means of evading standards of performance.



     The second alternative is to base standards of performance on the



firing of low nitrogen fuels.  This approach would require emission



controls on all new stationary gas turbines, but would effectively



preclude the firing of fuels other than those premium gaseous and distillate



fuels which turbines are now using.  Firing of heavy fuels would require



major breakthroughs in controlling the contribution of fuel-bound nitrogen



to NO  formation.
     A


     The third alternative is to include an allowance in the NO  emission
                                                               A


limit which is a function of the fuel-bound nitrogen level of the fuel fired.



This approach would require NO  controls on all new stationary gas turbines,
                              A


but would not restrict new gas turbines to firing premium gaseous and



distillate fuels.   Thus, new stationary gas turbines would not be penalized



for firing heavy fuels, nor would there be any added impetus toward the firing



of heavy fuels in order to evade standards of performance.



     As discussed earlier, low sulfur fuels, such as premium distillate



fuel  oils or natural gas, are now being fired by nearly all stationary



gas turbines.  These premium fuels are being burned primarily because the



increased maintenance costs associated with firing heavy fuel oils are
                                  8-29

    -------
greater than the savings that would be realized by buying these less



expensive heavy or residual fuel oils.  Over the next five to ten years,



however, as oil prices continue to escalate, the price differential



between premium Distillate fuel oils and heavy fuel oils will probably



increase and the economic incentive to burn the premium fuel oils will



probably become marginal.  It is also possible, however, that there could



be limited supplies of premium distillate fuel oils due to declining



production of oil and natural gas in the United States, increased demands



for these premium fuels by users other than gas turbines which cannot



utilize heavy or residual fuel oils, and the uncertainty of additional



crude oil supplies in the world energy markets.   In the event of limited



supplies, many new gas turbines would probably be  designed to fire



residual or heavy fuel  oils.  Consequently, in order to provide gas



turbine owners and operators the flexibility to fire either premium



or heavy and residual fuel oils, but  to ensure that standards of



performance add no impetus toward the firing of heavy fuel oils as a



means of evading standards, alternative three  is  selected for standards



of performance limiting NO  emissions from  stationary gas turbines.
                          A


     An allowance in  the NO  emission limit dependent on fuel-bound
                           A


nitrogen level with no upper limit  on emissions,  however, could permit



extremely high NO  emissions when firing some  very high  nitrogen-containing
                 A


fuels.  Thus,  it is essential  that  restraints  be  placed  on  such an emission



allowance.  Using the data presented  in Figures 8.3,  8.4 and 8.5,  a  fuel-



bound nitrogen allowance can be  developed  that allows approximately



50 percent  availability  of the  heavy  fuel  oils.   This corresponds  to



a fuel-bound nitrogen content  of about  0.25 weight percent.  Referring



to Figure 8.5, firing a  fuel with 0.25 weight  percent fuel-bound  nitrogen
                                   8-30

    -------
increases controlled NO  emissions by about 50 ppm.  Consequently, a fuel-
                       A


bound nitrogen NO  emission allowance based on a straight line approximation
                 X


of Figure 8.5, with a maximum allowance of 50 ppm, is selected for standards



of performance.



     The numerical NO  emission limit, therefore, is a function of thermal
                     A


efficiency of the gas turbine and fuel-bound nitrogen level of the fuel



fired as follows:
                        NO  = [0.0075 (E) + F]
                          A
     where:
     NO  = allowable NO  emissions (percent by volume at 15 percent
       A               X


           oxygen).



     E   = the efficiency adjustment factor:



                     14.4 kilojoules/watt-hr
           LHV heat input per unit of power output



     F   = the fuel-bound nitrogen allowance:



           Fuel-Bound Nitrogen

           (percent by weight)                   £


                N < 0.015                        0


             0.015 < N < 0.1                 0.04 (N)


             0.1 < N < 0.25           0.004 + 0.0067 (N-0.1)


                N > 0.25                       0.005
     The effect of ambient atmospheric conditions on NO  emissions from
                                                       X


stationary gas turbines is substantial.  Large changes in relative humidity,



for example, can cause NO  emissions to vary by a factor of 2 or more.
                         A


In order to insure that standards of performance are enforced uniformly,



therefore, the effect of ambient atmospheric conditions on NO  emission
                                                             X


levels needs to be taken into account.  The following equation to correct





                                   8-31

    -------
measured NO  emissions to ISO ambient atmospheric conditions was derived
           A


by extracting the common elements from several ambient correction factors



proposed by gas turbine manufacturers.  This correction factor, therefore,



represents the g,,ieral effect of ambient atmospheric conditions on NO
                                                                     A


emissions.  Consequently, this correction factor, or an alternative



factor as discussed below, will be used to adjust measured NO  emissions
                                                             A


during any performance test to determine compliance with the numerical



emission limit.
     N0  MNO    ,(,.eobs- 0.00633)


       x      xobs  Kobs




     where:



     NO     = emissions of NO  at 15 percent oxygen and ISO standard
       X                     X


              ambient conditions.



     NO     = measured NO  emissions at 15 percent oxygen, ppmv.

       xobs              x


     P  f     reference combustor inlet absolute pressure at  101.3



              kilopascals  (1 atmosphere) ambient pressure.



     P .    = measured combustor inlet absolute pressure.



     H .    = specific humidity of ambient air.



     e      = transcendental constant  (2.718).



     As an alternative, gas turbine manufacturers may elect to  develop



 custom correction factors  for adjusting measured NO  emissions  from
                                                   X


 particular gas  turbine models to ISO standard  ambient conditions  of



 pressure  (1 atmosphere), humidity  (60  percent  relative  humidity), and



 temperature  (288°K).  Some gas turbine manufacturers have proposed  ambient



 correction factors  which include variables such as fuel -to-air  ratios



 and  combustor temperatures.  These variables are difficult to measure and



 are  operating parameters which may vary widely due to factors other than
                                   8-32

    -------
ambient conditions.  For this reason, any custom correction factor must



be developed in terms of only the following variables: combustor inlet



pressure, ambient air pressure, ambient air humidity, and ambient air



temperature.  All such correction factors must be substantiated with



data and then approved by EPA for use in determining compliance with the



NO  emission limit.  The ambient correction factor will be applied to
  X


all performance tests, not only those in which the use of such factors



would reduce measured emission levels.



     As discussed in section 8.4, some delay is required before this emission



limit can be applied to small stationary gas turbines.  This delay in



effective date is to provide time for manufacturers to incorporate NO
                                                                     A


controls on their small production stationary gas turbine models.  It is



estimated that about three years delay in the effective date of the standard



for small stationary gas turbines would be required to allow manufacturers



time to incorporate and test wet controls on these gas turbines.  Some manu-



facturers have expressed optimism at being able to meet this emission



limit using dry controls if given about five years delay.  Since these



small turbines represent only about  10 to 15 percent of the total NO
                                                                    A


emissions from stationary gas turbines, the difference in environmental



impact of a three-year versus five-year delay would be quite small.



Additionally, a three-year delay would essentially force these manufacturers



to incorporate wet controls whereas  a five-year delay would provide the



flexibility to use wet controls or to develop and use dry controls.



Consequently, five years is selected as the delay period for implementation



of these standards on small stationary gas turbines.



     In selecting the size cutoff to differentiate between large and small



stationary gas turbines, consideration must be given to the purpose for




                                   8-33

    -------
the cutoff and the effect on competitive markets.   The purpose of the



cutoff is to differentiate between large gas turbines where wet controls



have been commercially demonstrated and small  gas  turbines where wet



controls, althorjii effective, have not been generally applied on a commercial



basis.  Consideration of the market data in chapter 3 reveals that there



are two major competitive markets for stationary gas turbines which can



be generally described as small gas turbines and large gas turbines.



The size range of 5000 to 10,000 horsepower essentially separates these



two markets.  All gas turbines above this range are manufactured by companies



which have developed wet control syterns for their stationary gas turbines.



The size cutoff, therefore, between small and large gas turbines is selected



as the upper end of this range.  Thus, large stationary gas turbines are



defined as those with heat input greater than 107.2 gigajoules per hour



(approximately 10,000 horsepower for a 25 percent efficient gas turbine).



     As discussed earlier in this chapter, S02 emissions  from a stationary



gas turbine depend on the sulfur content of the fuel fired.  The best



system of emission reduction,  considering costs, selected  for S0?



emissions was the firing of low sulfur fuels.  Selection  of the S0?



emission limit,  therefore, will be based on the use of this control



system.



     As also discussed earlier, nearly all  stationary gas  turbines  are



currently firing natural gas or premium distillate  fuel oil;  although



over the next five to ten years, some new gas turbines may fire heavy



or residual fuel oil for either economic reasons or  if a  shortage  in



supply of premium fuel oils should develop.  A fuel-bound nitrogen



allowance to permit  increased  NO  emissions has been  selected  to allow
                                A


turbines to burn approximately 50 percent of currently available heavy





                                  8-34

    -------
fuels.  To be consistent with the objective of the fuel-bound nitrogen



allowance, the SCL emission limit is selected as 150 ppm referenced to



15 percent CL.  This corresponds to a fuel sulfur content of approximately



0.8 percent by weight and would allow about 50 percent availability of



heavy fuel oils.



     The five-year delay of the NO  emission limit applied to small gas
                                  A


turbines (less than 10,000} to provide manufacturers time to incorporate wet



controls onto their turbines does not apply to the SO,, emission limit since



the control technique of burning low sulfur fuels is available to all turbines



at the present time.





8.7  MODIFICATION/RECONSTRUCTION



     A discussion of the modification and reconstruction regulations and



how they pertain to the gas turbine industry can be found in chapter 5.



Since few modified or reconstructed gas turbines are anticipated, the modifi-



cation and reconstruction regulations will have little impact.  Wet controls,



however, are as effective in reducing emissions of NO  from modified or
                                                     X


reconstructed gas turbines as from new gas turbines.  Thus, modified or



reconstructed gas turbines merit no special allowance in the standards of



performance.





8.8  SELECTION OF MONITORING REQUIREMENTS



     To provide a convenient means for enforcement personnel to ensure that



an emission control system installed to comply with standards of performance



is properly operated and maintained, monitoring requirements are generally



included in standards of performance.  For stationary gas turbines the most



straightforward means of ensuring proper operation and maintenance is to



monitor emissions released to the atmosphere.





                                   8-35

    -------
     EPA has established NO  monitoring performance specifications in
                           X


Appendix B of 40 CFR Part 60 for large industrial  sources  with well  developed



velocity and temperature profiles.   Stationary gas turbines, however, do not



have well develo,cd velocity and temperature profiles in all cases.   Gas



stratification of the turbine exhaust, for example, makes  the location of



the sample point critical.  Also, since some turbines are  started remotely



from a central location, special systems and data  reporting procedures would



be necessary to start and maintain continuous monitors.



     Currently there are no NO  continuous monitors operating on gas turbines,
                              A


and resolution of these sampling problems and development of performance



specifications for continuous monitoring systems would entail a major develop-



ment program.  For these reasons, continuous monitoring of NO  emissions
                                                             A


from gas turbines will not be required by the new source performance standard.



     A means of ensuring operation of the water injection system used to



control NO  emissions from gas turbines is to monitor the water-to-fuel ratio
          A


being fed to the turbine.  Both water and fuel monitors are readily available



and are demonstrated technology for use on gas turbines.  Consequently, to



ensure operation of water injection systems, the standards for stationary



gas turbines will require continuous monitoring of the water-to-fuel



ratio where water injection is employed to comply with the NO  standard.
                                                              A


     Also, a means of ensuring the firing of fuels with the proper nitrogen



content to control NO  emissions caused by fuel-bound nitrogen is to
                     A


monitor the nitrogen content of the fuel being fired.  Consequently, any



owner or operator that uses the fuel-bound nitrogen  allowance to comply



with the NO  emission limit will be required by the  standard  to monitor
           X


the nitrogen content of the fuel.



     The continuous monitoring of SOp emissions will not be required by the



new source performance standard for the same reasons continuous monitoring
                                   8-36

    -------
of NO  emissions will  not be required.   A means of ensuring the firing
     A
of low sulfur fuels to control  SO^ emissions, however, is to monitor the
sulfur content of the fuel  being burned.   This is already a common
practice among gas turbine  users.   Consequently, to ensure the use of
low sulfur fuels by stationary gas turbines to comply with the SO-
emission limit, the standard will  require monitoring of the sulfur
content of the fuel.

8.9  SELECTION OF PERFORMANCE TEST METHODS
     Reference Method 20, "Determination of Nitrogen Oxides, Sulfur Dioxide,
and Oxygen Emissions from Stationary Gas Turbines," is selected as the
performance test method to  determine compliance with standards of performance
limiting NO  emissions for stationary gas turbines.  This test method is
           A
based on the EPA gas turbine field tests and on background data for
continuous monitoring system specifications  (Federal Register, October 6, 1975).
Reference Method 20 includes (1) measurement system design criteria,  (2)
measurement system performance specifications and performance test
procedures, and (3) procedures for emission sampling.  The performance
specifications include the span drift, zero drift, linearity check,
response time of the system, and interference checks.  This method
allows a choice of instruments and will provide reliable data if the
performance specifications  are met.  Appendix D, section 3 and Appendix G
give a full explanation of Reference Method 20.
     The Mobile Source and SAE test procedures were considered as possible
performance test methods but were rejected because they specified the use
of particular types of instruments, rather than design criteria and performance
specifications of the measurement system.  Both the SAE and Mobile Source
                                  8-37

    -------
test methods, however, are acceptable alternative methods,  if the selected
                                             I
instrument models are capable of meeting the performance specifications
of Reference Method 20.
     As mentions earlier in this chapter,  the NO  emissions measured by
                                                 A
Reference Method 20 will be affected by ambient atmospheric conditions.
Consequently, measured NO  emissions will be adjusted during any performance
                         A
test to determine compliance by the following equation or by custom equations
developed by turbine manufacturers, owners  or operators and approved by the
Administrator.

     NO  = (NO    ) (/^)°'5 e19 
    -------
determined by determining the sulfur content of the fuels being used by



the gas turbine.  Sulfur content of the fuel will be determined using



ASTM D2880-71 for liquid fuels and ASTM D1072-70 for gaseous fuels.
                                  8-39

    -------
                              References



1.   Responses from 17 electric utilities  submitted  by  Baruch,  S.B.,  Edison



    Electric Institute,  to K.R.  Durkee,  EPA.   October  1975-January  1976.
                                   8-40

    -------
            APPENDIX A.  EVOLUTION OF THE SELECTION OF THE
                         BEST SYSTEM OF EMISSION REDUCTION
INTRODUCTION
     The study to develop proposed standards of performance for new
stationary gas turbines began in 1971.  In the course  of the program, a
literature search was conducted and contacts were made with practically
all domestic manufacturers of stationary gas turbines, several foreign
manufacturers, a number of users (electrical utilities and pipeline
companies), technical societies, trade associations, two trade journals
and several air pollution control districts.
     EPA personnel met with many of these interested parties on numerous
occasions and exchanged much correspondence and countless telephone
calls.  The dates and locations of the meetings are documented in Table A-l,
but the letters and telephone calls were too numerous to catalog.
     Considerable data was obtained from published reports and trade
publications.  Direct contacts with engine manufacturers also provided
much of the data on emissions from uncontrolled engines, on control
technologies, and on emissions from controlled engines.  Additional
information regarding stationary applications, control technologies,
and their costs were received from manufacturers in response to official
requests for data by the Director, Emission Standards and Engineering
Division, U.S. EPA.  These requests were sent to manufacturers under
                                A-l

    -------
the authority specified in Section 114 of the Clean  Air Act.   Most  of
the data on the effectiveness of the various  dry  controls  for  NO
                                                                /\
came from labor. :ory experiments at the manufacturer's  plants.   Much
of the test data on wet controls for NO , however,  is  from field tests
                                       A
by manufacturers and users.  For many of the  field  tests the control
equipment was a temporary installation and was utilized only for the
duration of the tests.  EPA also conducted tests  of  gas turbines at
two electrical utilities to validate the test methods  and develop
corroborating data.
     The direct contacts were supplemented by visits to the manufacturers
listed on Table A-2.  The purpose of these visits was  to obtain more
information than possible over the telephone  or by  letter concerning
the status of their R&D efforts 1n emission reductions, their  experience
with the commonly proposed control technologies,  their estimates of the
cost and time required to incorporate such controls  on their engines,
and the importance of the stationary market to them.
     Standards Support Documents were prepared and  standards with slight
variations were recommended by the Industrial Studies Branch  (ISB),
ESED, to the National Air Pollution Control Techniques Advisory Committee
(NAPCTAC)'on three occasions:
                         February 21, 1973
                         May 30, 1973
                         January 9, 1974
The primary pollutants of concern at that time were S02> NOX,  CO and
visible emissions.
                                    A-2

    -------
      After the January 9, 1974, meeting with the NAPCTAC, several meetings
were held with the industry.  New issues were brought up and considerable
time was spent in obtaining technical and economic data to resolve them.
In August 1974, a completely new Standards Support Document was completed.
In October 1974, final preparation of the wording of the regulation
and its preamble was begun.  After some initial work, the project was
delayed several months because of higher priority work.  During this time
period, EPA expanded the requirements for a Standards Support Document
to include inclusion of an abbreviated environmental impact statement.
This required the entire document (now called a Standards Support and
Environmental Impact Statement) to be rewritten, a project which was
begun in September 1975.
      Since more than a year had passed since the last Standards Support
Document was prepared (August 1974), several manufacturers and users
of gas turbines were contacted and recent literature was reviewed to determine
if any major changes had occurred either in control techniques or the
market.  There were several, such as:
      A.  Fuels were no longer scarce because the oil embargo had ended.
      B.  The trend to burn crude and residual oils in gas turbines had
been reversed.
      C.  Several manufacturers had aggressively^pursued the development of
dry controls (via changes in the combustor geometry) to reduce peak flame
temperatures and duration and thereby reduce NO  formation.
                                               /\
      D.  Many power companies had installed water treatment and injection
systems for controlling NO  emissions from turbines.  When the last document
                          y\
was written there were few installations using water injection, over 70
such turbines are now known.
                                  A-3

    -------
      E.  Market conditions for gas turbines had changed.
      In light of these changes;
      1.  Eight letters were sent out, under the authority of section
114 of the Clean Air Act, to obtain technical and economic data from
gas turbine users and manufacturers.
      2.  The same request was sent to two pipeline companies, two
manufacturers of gas turbines, and two industry associations who
volunteered to assist us by supplying information.
      3.  A contractor was engaged to obtain current information on
economic and market data.
      The information resulting from these efforts is incorporated into
the body of this report and the data are summarized in Appendix C.
                                    A-4

    -------
                      Table A-l.   MEETING RECORD

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Date
1/23/72
4/12/72
4/12/72
8/17/72
10/12/72
10/27/72
10/31/72
12/13/72
12/15/72
3/26/73
4/12/73
4/26/73
5/17/73
6/13/73
6/20/73
8/9/73
9/6/73
18
9/27/73
   Company or Association
Institute of Gas Technology
Westinghouse Electric Corp.
Turbo Power and Marine Systems,
Pratt & Whitney Aircraft
San Diego Air Pollution
Control District
Turbodyne Corporation
General Electric Company
Turbo Power and Marine Systems,
Pratt & Whitney Aircraft
Turbodyne Corporation
Turbo Power and Marine Systems
General Motors Corporation
Apollo Chemical
General Electric Company
Southern California Gas Co.
Columbia-Willamette Air
Pollution Authority
General Electric Company
Westinghouse Electric Corp.
Oregon Dept. of Environmental
Quality (OEQ)
Bell Labs
Meeting Location
 Durham, N.C.
 Durham, N.C.
 Durham, N.C.

 San Diego, Cal.
 Durham, N.C.
 Durham, N.C.
 Durham, N.C.

 Durham, N.C.
 Durham, N.C.
 Durham, N.C.
 Durham, N.C.
 Durham, N.C.
 Durham, N.C.
 Portland, Ore.
 Durham, N.C.
 Durham, N.C.
 Portland, Ore.

 Durham, N.C.
                                  A-5

    -------
19
20
21
22
13
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Pate
11/16/73
11/28/73
3/22/74
5/17/74
6/10/74 -
6/12/74
7/2/74
8/22/74
3/29/74
8/30/74
10/3/74
10/25/74
11/7/74
1/10/75
1/10/75
6/10/75
8/19/75
10/14/75
1/21/76
2/18/76
3/4/76
3/9/76
8/10/76
Company or Association
General Electric Company
Florida Power & Light Co.
American Gas Association
Ethyl Corporation
Air Pollution Control Assoc.
Annual Mtg.
General Motors, Detroit Diesel
Allison Division
General Electric Company
Solar Division of International
Harvester
American Nat'l Standards Inst.
Ai research Man. Co. of Arizona
Solar Division of International
Harvester
Westinghouse Electric Corp.
General Electric Company
Turbo Power & Marine Systems
American Gas Association
General Electric Company
Engine Manufacturers Association
Solar Division of International
Harvester
Edison Electric Institute
Turbo Power & Marine Systems
General Electric Company
NAPCTAC Meeting
                     Meeting. Location
                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C
                       Denver, Colo.

                       Durham, N.C.

                       Durham, N.C.
                       Durham, N.C.

                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.

                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       „Durham, N.C.

                       Durham, N.C.
                       Durham, N.C.
                       Durham, N.C.
                       Chicago,  II].
A-6

    -------
                 Table A-2.   SURVEY & INSPECTION TRIPS
 Date               Company

6/6/72      General  Electric Company

6/7/72      Bell  Labs

6/8/72      Turbo Power & Marine
            Systems

6/9/72      Westinghouse Electric
            Corp.

8/16/72     Airesearch Manufacturing
            Co.  of Arizona

8/17/72     Solar Div. of International
            Harvester

8/18/72     San  Diego Gas & Electric
            Company

2/15/73     Exxon Chemical Co.

2/16/73     Exxon Chemical Co.

3/20/73     Naval Air Rework Facility

6/11/73     Carolina Power & Light
            Company
       Location

Schenectady, N. Y.

Murray Hill , N. J.

Farmington, Conn.


Philadelphia, Pa.


Phoenix, Arizona


San Diego, Cal.


San Diego, Cal.


Baytown, Texas

Baton Rouge, La

Jacksonville, Fla.

Goldsboro, N. C.
6/13/73     Portland General  Electric     Portland, Oregon
6/13/73     Portland General  Electric     Salem, Oregon
4/9/74      General Electric Company

6/6/74      Detroit Diesel  Allison
            (GM)
Schenectady, N. Y.

Indiannapolis, Ind.
Jype of
Company

  Mfr

  User

  Mfr


  Mfr


  Mfr


  Mfr


  User
  (Utility)

  User

  User

  User-

  User
  (Utility)

  User
  (Utility)

  User
  (Utility)

  Mfr

  Mfr
                                 A-7

    -------
                                Appendix B
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)

1.  Background and Description of
    Proposed Action

    Summary of Proposed Standards
    Statutory Basis for the Standard


    Facility Affected


    Process Affected
    Availability of Control
    Technology

    Existing Regulations at State
    or Local Level
 Location Within the Standards
   Support and Environmental
 	Impact Statement	
The standards are summarized in
chapter 1.

The statutory basis for the standard
is given in chapter 2.

A description of the facility to be
affected is given in chapter 3.

A description of the process to be
affected is given in chapter 3.

Information on the availability of
control technology is given in chapter 4.

A discussion of existinq regulations
on the industry to be affected by the
standard is included in chapter 3,
section 2.3.5.
2.  Alternatives to the Proposed
    Action

    a.  Postponing Action

    Environmental  Impacts



    b.  Dry Control Alternative

    Environmental  Impacts



    Costs
Environmental effects of delaying
the standards are discussed in
chapters 6 and 8.
The environmental impacts associated
with this alternative are discussed
in chapter 6.

The costs of a dry control alternative
are discussed in chapter 7.
3.  Environmental  Impact of
    Proposed Action

    Air Pollution
The air pollution impact of the standards
is considered in chapter 6, section 6.1.
                              B-l

    -------
                          Appendix  B  (continued)
Agency Guidelines for Preparing
Regulatory Action Environmental
Impact Statements (39 FR 37419)

Water Pollution
Solid Waste Disposal
Energy
Location Within the Standards
  Support and Environmental
	Impact Statement	

The water pollution impact of the
standards is discussed in chapter 6,
section 6.2

The solid waste disposal impact of the
standards is discussed in chapter 6,
section 6.3.

The energy impact of the standards is
considered in chapter 6, section 6.4.
                                  B~2

    -------
               APPENDIX C.  EMISSION TEST DATA SUMMARY

C.I  INTRODUCTION
     This appendix summarizes the emission source test data cited in the
main body of this document.  It describes the tested facilities (gas
turbine size, type, operating conditions, and the exhaust gas stream
characteristics), reviews the testing methods used and summarizes the
results of the emissions measurement tests.
     Facilities are identified by the same coding used in the main body
of the document.  For example, Table 3 summarizes the emission source
test data obtained from facility C.
     A variety of sampling and analytical techniques were used to develop
the data contained in this appendix.  Each of these techniques is
discussed briefly in the next section.
     During the development of standards of performance for stationary
gas turbines,  the major gas turbine manufacturing companies  in the
United States were contacted, along with those electric "utility companies
known to have experience with the control of emissions from gas turbines.
       s
Those companies which manufacture most of the turbines sold in this country
were then visited as were a number of the electric utility companies
who operate gas turbines.
     An evaluation of emission control technology (as it applies to gas
turbines) was also undertaken by means Of a literature survey, personal
contacts and plant visits.  A great deal of research on combustion
                               C-l

    -------
modifications is underway and,  as  discussed in  Chapter 4,  has  already



demonstrated considerable success  in reducing NO  emissions  as evidenced
                                                y\


by rig tests anH tests of prototype engines.   These combustion modifications,



however, have not yet been applied to production engines  to  the extent of



achieving reductions in NO  emissions as great as those that have been
                          A


demonstrated through the injection of water or steam directly into the gas



ti -oine combustors.  EPA tested three turbines operated by one utility which



used water or steam injection.   The data listed in the tables of section C.5



are emission source test data reported to us by utility companies, turbine



manufacturers, and local pollution control agencies, as well as the



results of the EPA tests.



C.2  DESCRIPTION OF EMISSION TEST METHODS



     The test methods used to obtain the emission data summarized in



this appendix are summarized below in two parts.  The sampling procedures



have been identified by letter codes and the analysis techniques by numbers.



A test method can consist of any letter-digit combination.  Test method Al,



for example, consists of sampling method A and analysis method 1.



SAMPLING PROCEDURES



     A.  A movable probe assembly was used for sampling.  The assembly  is



         comprised of two adjustable single point probes, one to survey the



         vertical and the other to survey the horizontal radii of the



         engine  tailpipe.  Each probe is center-positioned  in each of five



         equal annulus  areas with a total of 10 points being  sampled.



     B.  A fixed probe  technique was used for sampling.  The  probe



         incorporated three sampling points in each of four parallel



         branches which were manifolded to a single sample  line.
                                 C-2

    -------
C.   The sampling technique described in the Los Angeles Air
    Pollution Control  District Source Testing Manual  was used.
D.   Two fixed probes located at right angles to each  other and
    perpendicular to the gas stream.  Each probe contained 11
    sampling ports.   Sample lines were manifolded together to
    provide an average gas sample.
E.   Three two-litre grab samples were taken during each test for
    NO  determination with continuous sampling for C09 and 0?.
      X                                              L.      £
F.   The single point sampling techniques described in the Los
    Angeles Air Pollution Control District Source Testing Manual
    were used.
G.   Samples were obtained by traversing the gas turbine stack with
    a single port probe.  The number of sampling points was selected
    in accordance with EPA Method 1, entitled, "Sampling and Velocity
    Traverses for Stationary Sources", which was published
    December 23, 1971.
H.   Samples were extracted at four pressure probes located between
    the compressor and power turbines.  Sample lines  were manifolded
    together to provide an average gas sample.
J.   Three stationary probes were located at each of two sampling
    stations, one upstream of the exhaust silencing baffles and the
    other downstream.
K.   A fixed, single-point probe was used for sampling.
L.   Two gas-sampling techniques were used, an annular traversing
    probe and a fixed sampling rake.  The traversing  probe contained
    four radial sampling ports which were normally manifolded together.
    The probe was centered in each of five equal annul us areas  with a
    total of 20 points being sampled.  The fixed rake contained five
                             C-3

    -------
         sampling  probes, each  containing  four  radial sampling ports
         connected to a single line.   The five probes  (20 sampling
         pointij were normally manifolded together  to  provide a  single
         sample.
     M.   A traversing rake was used  for sampling.   It  was comprised  of
         five probes  which radiated  out from the  engine  exhaust  center
         line and  was rotated  through 12  traverse positions  for  each test.
         Each probe contained  five ports, of which  three were normally  used
         for gas sampling - for a total of 180 samples per traverse.
         Usually,  the lines from each port in  a probe  were manifolded
         together  to  provide an average gas sample  for each  probe.
     N.   Samples were obtained by sampling at  seven radial locations around
         the stack using a five port probe, for a total  of 35 sampling  ports.
         Sample lines from each port were manifolded together to provide
         an average sample for each  probe position.  The number  of sampling
         points was determined in accordance with EPA Method 1,  entitled,
         "Sampling and Velocity Traverses for  Stationary Sources", which
         was published December 23,  1971.
ANALYSIS TECHNIQUES
     1.   Chemiluminescent analyzer with a thermal converter  was  used to
         convert N07 to NO for total NO  , a non-dispersive  infrared detector
                   c.                   /\
         (NDIR) was used for CO and C02 and a   flame ionization  detector  (FID)
          measured HC emissions.
     2.   The phenoldisulphonic acid  (PDS) and  Saltzman methods were  used  for
         analysis  of NO  , a gas chromatograph  (GC)  was used  for  CO
                       /\
         determination and an  Orsat analyzer was  used to measure (L and COp
         concentrations.
                                 C-4

    -------
     3.   The PDS and Saltzman methods  were  used  for analysis  of NO  ,
                                                                 A


         CO was estimated with a Mine  Safety  Appliance  (MSA)  indicator



         tube,  hydrocarbons  (HC) levels  were  determined with  an FID and



         Op and Op were measured with  an Orsat analyzer.



     4.   PDS and electrochemical methods were used to determine NO
                                                                 J\


         concentrations, an  NDIR measured CO  and HC emissions,  0? was



         measured by Orsat and with a  process monitor,  and  COp  was



         measured by Orsat.



     5.   PDS and electrochemical techniques were used for analysis  of



         NO .   An electrochemical analyzer  was used for CO  determination,
           J\


         an FID measured HC in the exhaust  gases, 02 and C02  levels were



         determined using GC techniques  and smoke was measured  by ASTM



         Smoke Spot method D-2156.



     6.   NDIR and Ultraviolet Analyses were used to measure NO, and NOp,



         and NDIR was used to measure  CO.  Op was calculated  from the



         fuel  air ratio.



     7.   NO  was determined using PDS, electrochemical, chemiluminescent
           /\


         and integrated lead dioxide techniques, CO was measured by NDIR,



         HC by FID, and Op by polargraphic  and paramagnetic techniques.



C.3  DESCRIPTION OF FACILITIES



     A.   A single shaft gas turbine with a  rated output of 0.03 megawatt



(40 hp).  The production (Al) turbine  had no  emissions  controls. The



development (A2) turbine utilized vaporizing  combustors for emissions  control




Test Method Al  was used and data were  provided by the manufacturer.   The



turbine was tested in idle mode and at rated  output. Fuel  used was aviation



kerosene.
                                   C-5

    -------
     B.  A single-shaft gas turbine with a rated output equivalent to
0.158 megawatts (212.5 hp).  These turbines had no emissions controls.
They were tested over the operating load range using method Al  and data
were provided by the manufacturer.  The gas turbines in Test series 1
were operated on Jet A-l fuel and the unit in test 2 was operated on
aviation kerosene.
     C.  An advanced engine test rig for a single-shaft gas turbine with
an output of about 0.158 megawatt (212.5 hp).  A number of dry control
techniques were tested over the operating load range, singly and in
combination, using test method Al and data were provided by the manufacturer.
Units were operated on aviation kerosene and natural gas during the testing.
     D.  A combustor test rig for a single-shaft gas turbine with a
rated output equivalent to 0.158 megawatt (212.5 hp).  This combustor was
tested without NO  controls and also with staged fuel combustion.  It was
                 ^\
tested over the operating load range using method Al and data were supplied
by manufacturer.  Fuel used for these tests was not specified, but was
probably aviation kerosene.
     E.  A single-shaft gas turbine with a rated output of 0.25 megawatt.
It had no emissions controls.  Test method Bl was used and data were
provided by the manufacturer.  Operation was at rated output and the fuel used
was aviation kerosene.
     F.  A regenerative-cycle two-shaft gas turbine with a rated base load
output of 0.20 megawatts.  This turbine incorporates a short residence time
primary zone to reduce NO  formation.  The test method is unknown.  The data
                         /\
were supplied by  the manufacturer and are for operation at rated output.
Fuels used were #2 distillate and natural gas.
     G.  A single-shaft aircraft turboprop engine which is aerodynamically
identical to facility H.  The Gl engine has a rated shaft output of 665
horsepower (.497 MW), had no N0x controls, and was tested over the operating
                                C-6

    -------
load range using method Al.   The G2 engine is rated at 690 horsepower
(.51 MW) and was tested at 100% of design capacity with water injection,
also using method Al.   Data  for Gl was developed for EPA by the manufacturer
and G2 data was supplied by  the manufacturer.  Gl  tests 1  and 2 were
run with aviation kerosene and JP5, respectively.   The fuels used for most
of the G2 tests were not specified.
     H.  A single-shaft industrial gas turbine with a rated output of
0.51 megawatts (690 hp).  The HI engine was tested over the operating
load range using method Al  and data was provided by the manufacturer.
Tne HI test data represents  an average of six tests on units operating on
natural gas (test series 1)  and 11 on DF-2 (test series 2).  For H2, a
number of dry control  techniques were tested, singly and in combination,
during operation on DF-2.
     J.  A combustor test rig for a single-shaft gas turbine with a
rated output of 0.51 megawatts  (690 hp).  A number of dry control techniques
were tested, singly and in combination, over the operating load range.
Test method Al was used and data was provided by the manufacturer.  Fuel
used was DF-2.
     K.  A single-shaft or two-shaft gas turbine with a rated output of
1100 horsepower (0.75 - 0.80 MW continuous output).  These turbines had no
emissions controls.  Facility Kl was tested over the operating load range
using test method Cl plus polargraphic analysis for 02 and chemical absorption
for COp and O^.  The data were provided by a user.  Facility K2 test series,
15  units tested on kerosene (test  1) and 9 units tested on natural gas (test 2),
were conducted at rated output  using test method Bl, and the data were
provided by the manufacturer.
                                   C-7

    -------
     L.  Single-shaft gas turbine with a rated output of 1.0 megawatts.
This turbine had no emissions controls.  The test method is unknown.  It
was tested over the operating load range on #2 distillate fuel and data
were provided by the manufacturer.
     M.  Single-shaft or two-shaft gas turbines with a rated peak output
of 2.5 Ml^or 2710 hp continuous.  Three turbines were tested on kerosene
(test series 1) and 8 on natural gas (test series 2).  They had no emissions
controls.  The turbines were tested at rated capacity using test method Ml,
and data were provided by the manufacturer.
     N.  A production prototype gas turbine with a rated output of 2.5
megawatts.  It had no NO  controls but did use an air atomizing combustor
                        /\
for smoke reduction.  It was tested over the operating range using test
method Ml and data were provided by the manufacturer.  Fuels used were
natural gas, kerosene and #2 distillate oil.
     0.  A regenerative cycle two-shaft development turbine with a rated
output of 2.5 megawatts.  This turbine utilized a lean primary zone for
NO  control.  It was tested at no load on natural gas and at rated output
  /\
on all fuels.  Test method Ml was used and data were provided by the
manufacturer.  Fuels used were natural gas, kerosene, naptha and #2 distillate
oil.
     P.  A single-shaft gas turbine with a rated output of 2.5 megawatts.
Facility PI had no NOV controls but the combustor and fuel nozzle had been
                     A
modified for smoke and particulate emissions control.  It was tested at no
load on #1 distillate oil and at rated load on both #1 distillate and natural
gas.  The test method used is unknown and data were provided by the manufacturer.
Facility P2 was tested before and after combustion modifications were made
to reduce particulate emissions.  It was tested at no load and at rated output
on distillate fuel using test method Cl and data were provided by a user.
                                 C-8

    -------
Chemical absorption was also used for C02 and 02 measurement, and polargraphic



analysis for 02.



     Q.  A combustor test rig for a single-shaft gas turbine with a rated



output of 2.5 megawatts.  Water injection was used for NO  control.  Water
                                                         J\


was injected into the inlet air stream of a production combustor (test 1)



or into the combustor itself (test 2) at several flow rates during operation



at full rated load.  The test method used is unknown.  Test data were



provided by the manufacturer.  Fuel used was #1 distillate.



     R.  A combustor test rig for a single-shaft gas turbine with a rated



output of 2.5 megawatts.  An R & D combustor was equipped with variable



geometry and staged combustion for NO  control.  It was tested at no
                                     A


load and at rated output on # distillate fuel.  The test method used is



unknown.  Data were provided by the manufacturer.



     S.  A gas turbine with a base load rating of 6.0 megawatts (7.5 megawatts



peak).  Prevaporization of the liquid fuel is utilized to reduce emissions.



The test method is unknown, except that PDS was used for NO  analysis.  Data
                                                           J\


were provided by the manufacturer and are for the turbine operating on



distillate oil at rated base load capacity and on natural gas and distillate



oil at peak load output.



     T.  A two-shaft, free-turbine (aircraft type) gas turbine with a



rated output of 10.3 megawatts (13,900 horsepower).  It had no NO  emissions
                                                                 J\


controls.  Facility Tl was tested at rated output oh natural gas and



distillate oil and data were provided by a user.  The test method is unknown.



Facility T2 was tested over the load range using method Dl (less C02 measurment)



and data were provided by the manufacturer.  Fuels used for both series of



tests were natural gas and distillate oil.





                                  C-9

    -------
     U.  A two-shaft, free turbine (aircraft type)  gas  turbine with a
rated peak load output of 13 megawatts (13 MW base).   Facility U3 had
no NO  centre".  Water injection was used for NO  control  on Facilities
     X                                          X
Ul and 112.  Water was injected into the combustor at several flow
rates during operation at rated and peak loads.   Test method F2 was used
and data were provided by a user.  Fuel used was natural  gas and, for one
cest, #2 distillate oil.
     V.  A single-shaft gas turbine with a rated peak load output of 17.2
megawatts.  Facilities VI and V2 were equipped with an atomizing air-fuel
injection system to reduce visible emissions and utilized water injection
for NO  control.  Tests were run at spinning reserve and at about peak
output using natural gas and #2 distillate oil.   Facility VI was tested
using test method E3 and data were provided by a user.  Facility V2 testing
was accomplished using test method G7, with some tests performed using test
method F7.  Facility V3 was tested with no controls, with dry controls,
with water injection, and with a combination of wet and drv controls.
Tests were run at peak output using natural gas and #2 distillate oil.  Test
method F4 was  used and data were provided by the manufacturer.
     W.  A two-shaft, free-turbine gas turbine with a rated maximum peak
output of 21.3 megawatts.  Steam injection was used for NO  control.  Most
                                                          J\
of the tests in this turbine used test method G7, with a few tests using test
method F7.  Tests were  run in the spinning reserve mode and at peak output.
The fuel used  was JP-5.
     X.  A two-shaft gas turbine with  a rated peak output of 27.6 megawatts.
It had no emissions  controls.  Test method H6 was used plus Orsat for COp,
and data was provided by the manufacturer.  Fuels used were distillate oil
and methanol.  The turbine was tested  at  about 70% of rated capacity, and
results on methanol  fuel were extrapolated to the full load point.  The
fuel delivery  system was sized for distillate, which has about twice the
                                   C-10

    -------
heat of combustion of methanol, so full power output could not be achieved



with methanol fuel.



     Y.  A single-shaft gas turbine with a rated peak output of 32.8 megawatts.



Water injection was used for NO  control.  This turbine was tested using
                               A


test method J5 and data were provided by manufacturer.  Tests were run at



low load and at rated capacity using #2 distillate oil.



     Z.  A combustor test rig for a single-shaft gas turbine with an



unspecified power output.  A number of dry control techniuqes and water



injection were utilized for NO  control, singly and in combination, and tests
                              X


were run at rated output.  Conventional combustors were also tested, for



reference.  The sampling method used is unknown, but analysis method 5



was used.  Data were provided by the manufacturer.  Facility Zl and Z2



were both tested on #2 distillate oil and facility Zl was also tested on



natural gas.



     AA.  A  scaled combustor test rig for a single-shaft gas turbine of



unspecified  output.  Various dry control techniques and steam injection



were utilized  for NO  control.  Conventional combustors ware also tested,
                    /\


for reference.  Tests were run at rated output using test method Kl.  Also,



NOY was measured by NDIR and smoke by ASTM D-2156.  Data were supplied
  ^\


by the manufacturer.  Fuels used were natural gas and #2 distillate oil.



     BA.  A  scaled combustor test rig for a single-shaft gas turbine of



unspecified  output.  Catalytically-supported combustion was used for NO
                                                                       t\


control, with  a conventional combustor and a premix combustor included for



comparison.  Facility BA1 was tested using test method Kl.  Also, NO  was
                                                                    /\


measured by  NDIR and smoke by ASTM D-2156.  Data were supplied by the



manufacturer.  Facility BA2 was tested with an unknown sampling method



and analysis method 1, and data were provided by control device manufacturer.



A variety of fuels were used in the tests, including synthetic coal gas and



nitrogen-doped propane and distillate.




                                   C-ll

    -------
     CA.   An engine test rig for a dual  spool  axial  flow turbofan



aircraft engine producing 44,300 pounds  of thrust at take-off.   A number



of dry control .echniques were utilized  for NO  control  and a conventional
                                              rt


combustor was included for comparison purposes.  Test method L6 was used



plus NDIR for C02 and SAE ARP 1179 for smoke.   Data  were developed by the



manufacturer for NASA and tests were run at rated output on Jet-A fuel.



     DA.   A combustor test rig for a dual rotor high bypass ratio turbofan



aircraft engine producing about 50,000 pounds  of thrust at take-off.



Variable geometry and exhaust gas circulation  were utilized for NO  control.
                                                                  /\


Conventional combustor results are also  included for reference.  Test



method Ml was used, less HC measurement  and plus SAE All79 for smoke.  Data



were developed by the manufacturer for NASA.  Tests  were run over the



load range and are reported in the data  summary for rated output (take-off)



and 69%,  which is assumed equivalent to  base load for a stationary gas  turbine.



Fuel used was Jet-A.



     EA.   A single-shaft gas turbine with a rated base load output of 26



megawatts.  It had no NO  controls.  The sampling procedure is unknown,
                        /\


but analysis method 5 was used and data was provided by the manufacturer.



The turbine was tested over the load range  (up to 85% of base load) to



determine the effect of different fuels  on emissions.  Fuels used were



natural gas, #2 distillate, and a heavy distillate.



     FA.  A single-shaft simple cycle gas turbine with a rated output at



base load of 51.7 megawatts.  Water injection was used for NO  control.
                                                             A


Test method Nl was used (less hydrocarbon measurement) and Op was measured



using paramagnetic techniques.  The turbine was tested over the  load range



and data were supplied by the manufacturer.  Fuel used was #2 distillate.
                               C-12

    -------
     GA.  A single-shaft gas turbine with a rated output of 52.9
megawatts.  It had no NO  controls.  The sampling method is not known,
                        A
but analysis method 1 was used and data were provided by the manufacturer.
The turbine was tested at rated output on #2 distillate and on crude oil.
     HA.  A single-shaft simple cycle gas turbine with a rated base load
output of 61.5 megawatts on natural gas and 60.4 megawatts on #2 distillate
fuel.  Water injection was used for NO  control.  It was tested on distillate
                                      A
oil (HA1) and natural gas (HA2) at several water flow rates during operation
at close to rated output.  The sampling method is unknown, but analysis
method 1 was used, except that CO- was not measured.  Data were provided
by the manufacturer.
     JA.  A single-shaft regenerative cycle gas turbine with a rated
base load output of 58.6 megawatts.  It had no NO  controls.  The sampling
                                                 A
method is unknown, but analysis method 1 was used.  Also, 02 was measured
using paramagnetic techniques.  Tests were run at close to rated output
on #2 distillate oil, and data were supplied by the manufacturer.
C.4  TEST DATA CODING SYSTEM
NO., Controls
——x
     A.  Combustion Modification (Dry Controls)
         1.  Lean Primary Zone
           <
             a.  45% Primary air
             b.  60% Primary air (no secondary air)
         2.  Premix/Lean premix/Staged premix
         3.  Reduced residence time
         4.  Variable geometry
         5.  Exhaust gas recirculation
         6.  Staged fuel injection
                                C-13

    -------
          7.   Radial/axial fuel staging
          8.   Airblast/piloted airblast
          9.   V?~orizing  combustor/prevaporizing chamber
         10.   Pressure  atomizer
         11.   Lean  dome double annular
         12.   Staged  combustion
         13.   Swirl can
         14.   Swirl vorbix
     B.   Water Injection
          1.   Water added in  combustor primary  zone.
              M.  Methanol added  to  water.
          2.   Water added ahead of  combustor.
     C.   Combustor  Modification Plus  Water  Injection
     D.   Steam Injection
          1.   Water added in  combustor primary  zone.
          2.   Water added ahead of  combustor.
     E.   Catalytically  Supported  Combustion
FUELS
     DF-1        #1 distillate fuel
     DF-2        #2 distillate fuel
     Jet-A       Aviation kerosene
     JP-5        Jet  Kerosene
     K           Kerosene
     M           Methanol
     P           Propane
     NG          Natural  gas
     CG          Coal gas
     SCG         Synthetic  coal  gas

                                   C-14

    -------
Efficiency Correction Factor (EF)
     The efficiency factor (EF) is applied to the NO  emissions concentration
                                                    J\
(corrected to 15% 02) using the following formula:

          X = X1  (Y)
              10180
          Where:
          X = NO  concentration (@ 15% 0?) after correction factor applied
                /\                       C*
          X1 = NOX concentration @ 15% 02
          Y = Heat rate of test unit, in Btu/hph

     The purpose of this formula is to correct for the change in Op content
of the exhaust gas stream which is directly attributable to the greater
efficiency of the test unit, compared to the reference unit (which has a
heat rate of 10,180 Btu/hph).
                                     C-15

    -------
                                 TABLE 1


                             Facility  Al and A2


                            Summary of  Results
Unit Type



Test Number



"est Date



% Rated Output  (0.03 mw)



NL  Controls
  x


Water-Fuel Ratio



Fuel



Stack Effluent


  Flow rate - Ib/sec



  Temperature - °F



  Water vapor -Vol. %
Production Gas Turbine
1A
5-17-71
1
0 102.8
IB
5-18-71
1
0 102.4
Development Gas Turbin
2A
o 100
f
0
'B
100
None A-9 None A-9
640   795
                — Jet A —
610  850
C02 - vol. % (wet) 1.72
02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrocjen Oxides Emissions (as NOj)
ppm (dry) 12.7
ppm @ 15% 00 /-,\
ppm & 15% 0| w/EFu;
Ib/hr 0.84
Carbon Monoxide Emissions
ppm (dry) 331
ppm @ 15% 0
2
Ib/hr ].34
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% 02
Ib/hr .121
2.32
17.9
•*•
21.6
42
42
0.145
. 170
347
0.69

? • '
.015
1.74 2.20
18.0
*
19.2 26.2
53
53
1.31 0.195
235 333
666
0.97 1.50

-
.017 .017
                                                           28.8  36.9  43.5  46.9

                                                           28.8  36.9  43.5  46.9
                                                           751   291  536
                                               595
                                                            119    11.3   16.5    11.6
  NOTES:   (1)  Efficiency factor of no benefit.
                                    C-16

    -------
                               TABLE 2
                              Facility B
                         Summary of Results

Unlt Type                                        Production Gas  Turbine
Test Number                              1A^       1B^       2A^
Test Date                                9/73 thru  3/75             5/75
% Rated Output (0.158 equiv. mw)
NO Controls
/\
Water- Fuel Ratio
Fuel
Stack Effluent
Flow Rate - Ib/sec
Temperature - F
C02 - Vol. % (dry)
02 - Vol. % (dry)
Visible Emissions - % opacity
- SAE smoke
Nitrogen Oxides Emissions
ppm (dry)
ppm (? 15% 02
ppm @ 15% Op @ EF^ '
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm @ 15% 02
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% 02
Ib/hr
0 90 0 90
None None

Jet-A Jet-A

4.14 3.08 4.18 3.14
534 1163 525 1160



45.3 33.8 38.2 29.7


46.1 54.7 50.2 57.6
46.1 54.7 50.2 57.6
0.37 0.93 0.39 0.97


1282 323 1163 306
6.26 3.36 5.46 3.14


469 6.4 5.46 3.3
1.30 0.04 1.47 0.02
                                     C-17

    -------
                              TABLE 2
                             Facility B
NOTES:
1.   Average of tests on 6 units
2.   Single tests
3.   Efficiency factor of no benefit
                                   C-18

    -------
Unit Type
Test Number
Test Date
% Rated Output (0.158
NO  Controls
  /\
Water-Fuel  Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
                o
  Temperature -  F
  Water vapor - Vol.  %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions -  % opacity
Ni trogen Oxides Emjssjpns
  ppm (dry)
  ppm @ 15% 02
  ppm @ 15% 02 9 E
  Ib/hr
Carbon Mpnoxj de Emi ssions
  ppm (dry)
  ppm @ 15$ 02
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 09
                                        TABLE  3
                                       Facility  C
                                   Summary  of  Results
 1A
IB
 0    100
   None

   Jet-A
223
223
 44
Advanced Engine Test Rig
2A    2B     3A    3B    3C
       0   100
         A-8

         Jet-A
   3D
             0   100     0   100
                A-2, A-4 & A-9
                Jet-A
N.G.
     182   199    136   180   104   140
     182   199    136   180   104   140
      93    36.6  361   120   184   65.9
  3.6  -
       3.5  13.6   16.9   6.4  11.9  4.9
Notes:
1.  Efficiency factor of no benefit
2.  Estimated output of engine not necessarily of test rig
                                        C-19

    -------
                                       TABLE  4
                                      Facility D
                                  Summary of Results
Unit Type
Test Number
Test Dete
% Rated Output (.158 mw
NOX Controls
Water-Fuel Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
 Visible Emissions - % opacity
Nitrogen Oxides Emissions
  ppm (dry)
  ppm @ 15% 02
  ppm @ 15% 02 @ EF")
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hyd rocarbon Emj s s i ons
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
                                                        Combustor Test Rig
                                                 1A      IB             2A
                                                 0     100
                                                    None

                                                    UNK

                                                 1.11    1.14
                                                 64.4      57.4
                                                 64,4      57.4
                                                453
327
                        2B
                0     100
                    A-6

                    UNK

                1.11    1.14
               74.3    54.7
               74.3    54.7
1286     537
                                                  5.9
  3.0
 405
8.0
Not"s:
1.  Efficiency factor of no benefit.
2.  Output of engine,not necessarily  test rig.
                                        C-20

    -------
                                        TABLE  5
                                       Facility E
                                   Summary of Results

Unit Type                                              Production Gas Turbine
Test Number                                                       l^1)
Test Date                                                    9/72 -  8/73
% Rated Output (0.20 mw)                                        100
NO  Controls                                                    None
  /V
Water-Fuel Ratio
Fuel                                                            Kerosene
Stack Effluent
  Flow rate - Ib/sec
  Temperature -  F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity                                   *5
Njtrggen, Oxides Emissjons
  ppm (dry)
  ppm @ 15% 0                                                    56
  ppm 9 15% 02 @ EF                                              56
  Ib/hr
Carbon Monoxide Emi s si ons
  ppm (dry)
  ppm @ 15% 02                                                  250
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Notes:
1.  Data  represents average of tests on 9 units
2.  Corrector factor of no benefit
                                       C-21

    -------
                                        TABLE  6
                                       Facility F
                                   Summary of Results
Unit Type
Test Number
Test Dete
% Rated Output (0.20 mw)
NO  Controls (Dry)
  y\
Water-Fi.al Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissions
  ppm (dry)
  ppm @ 15% 0-
  ppm @ 15% 02 0 EF
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Regenerative Cycle Stationary Gas Turbine
           1                   2
         100
           A-3

           DF-2
          17.7
           2

         115
         209
         173
          82
         149
100
  A-3

  N.G.
 17.7
  2 (est.)

 62
113
 94
 38
 70
                                        C-22

    -------
                                      TABLE  7
                                     Facility Gl
                                 Summary of Results
Unit Type
Test Number
Test Date
% Rated Output (.497 mw)
NO  Controls
  x
Water-Fuel Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissigns
  ppm (dry)
  ppm @ 15% 00
                                               Production Aircraft Turbo Prop Engine
               9 EF
                   (1)
  ppm @ 15%
  Ib/hr
Carbon Mongxjde Emissions
  ppm (dry)
  ppm 0 15% Op
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% Q?
  Ib/hr
                                                  1A      IB
                                                    6/1/71
                                                  0.2    91.2
                                                     None

                                                     Jet-A
                                                735
1050
                                                  3.1
                                                  0.15
                                                388      16.4
                                                         21
                                                  3.19    0.35

                                                549       5.9
                                                          7.7
                                                  2.58     .071
                2A        2B
                   4/30/71
                0.6      90.6
                    None

                    JP5
1080
3.45
16.4
66.5
88
88
2.30
3.64
16.1
16.6 119
147
147
0.24 4.00
              381        34.8
                         43
                3.34      0.71
               99
                0.31
   7.6
   9.3
   0.05
Notes:
1.  Efficiency factor of no benefit
                                         C-23

    -------
                                      TABLE   8
                                     Facility G2
                                 Summary  of  Results

Unit Type                                        Aircraft Turbo Prop  Engine
Test Nunber                          1A      IB     1C       2A      2B     3A      3B
Test Date
% Rated Output (.51  mw)                     100               100            100
NOX Controls                                B-2               B-l            B-1M
Water-Fuel Ratio                     0.5      1.0     1.5     0.5     1.0    0       0.5
Fuel                                        UNK               UNK            Jet  A
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  CO  - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions -  % opacity
  % change                         +40     +45     +48     -23    -34
Nitrogen Oxides Emissions
  ppm (dry)
  ppm @ 15% 02                                                          148     87.6
  ppm @ 15% 02 @ EF^                                                  148     87.6
  Ib/hr
  percent change                   -41     -41     -50      -36   -60      N/A    40
Carbon Monoxide Emissions
  ppm (dry)
  ppm @ 15% 02                                                           17.3  1069
  Ib/hr
  percent change
Hydrocarbon Emissions
  por (dry)
  ppm @ 15% 02                                                           8.5    489
  Ib/hr
  percent change                  negl.   negl.   negl.      00—

Note:
1.  Efficiency factor of no benefit.
                                        C-24

    -------
     TABLE  9
    Facility HI
Summary of Results
Unit Type
Test Number
Test Date
% Rated Output (0.51 mw)
NO  Controls
  A
Water-Fuel Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
                  - SAE smoke
Ni tnogen Oxides Emi ssIons
  ppm (dry)
  ppm @ 15% 0?
  ppm @ 15% 02 @ EF^3'
  Ib/hr
Carbon Monoxjde Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emi ss i ons
  ppm (dry)
  ppm @ 15% 02
  Ib/hr

Notes:
1.  Average of tests on 6 units
2.  Average of tests on 11 units
3.  Efficiency factor of no benefit.
             Production Industrial  Engine
          1A(D      1B(D         2A(2)      2B(2)
          1/21 thru  5/12/75     10/74 thru 3/75
          0         97.3          0        90
               None                    None
               N.G.
        518
        119
          1.05
         24.6
          0.12
919
  7.0
  0.15
  4.2
   .05
                    DF 2
521
                                 19
 147
  1.27
 35.5
  0.18
876
                        27
102
102
1.46
109
109
3.69
125
125
1.77
149
149
4.76
 21.1
  0.41
 17.8
  0.20
      C-25

    -------
                                     TABLE 10
                                    Facility H2
                                Summary of Results
Unit Type
Test Number                   1A      IB
Test Date
% Rated Output (0.51 mw)      0     100
NO  Cortrols                     None
  A
Water-Fuel Ratio
Fuel                             DF-2
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissions
  ppm (dry)
  ppm @ 15% 02              136
  ppm @ 15% 02 0 EF* '      136
  Ib/hr
Carbon Monox j de Emi s s i on s
  ppm (dry)
  ppm 0 15% 02              130
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm 0 15% 02
  Ib/hr
                                    157
                                    157
                                                 Production  Industrial  Engine
                                              2A      2B       3A      3B     4A      4B
                                              0     100
                                                  A-l

                                                  DF-2
127
127
126
126
                                     16.2   336
         51.2
                  0    100
                     A-8

                     DF-2
175
175
164
164
                         0     100
                          A-l  & A-8

                          DF-2
133
133
139
139
         93.8   22.5  231
                        44.8
Notes:
1.  Efficiency factor of no benefit
                                         C-26

    -------
                                      TABLE 13
                                     Facility K2
                                 Summary of Results
Unit Type
Test Number
Test Date
% Rated Output (0.80 mw)
NO  Controls
  x
Water-Fuel Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature -  F
  Water vapor - Vo.. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Mitrogen Oxjdes Emissions
  ppm (dry)
  ppm @ 15% 02
  ppm @ 15% 02 0 EF
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Production 1 & 2 Shaft Gas Turbines
    1
     (1)
9/72-8/73
  100
   None

   Kerosene
  860
   16.5
   52
   69
   69
  122
  163
    ,(2)
9/72-8/73
  100
   None

   Natural Gas
  860
   17.0
   34
   51
   51
   45
   68
Notes:
1.  Data represents average of tests on 15 units.
2.  Data represents average of tests on 9 units.
3.  Efficiency factor of no benefit.
                                         C-27

    -------
                               TABLE 11
                              Facility  J
                          Summary of Results

Unit Type                                           Combustion  Rig
Test Number                  1A     IB      2A      2B     3A      3B      4A      4B
Test Date
% Rated Output (0.51  mwr2'    0     100    0      100      0      100      0      100
NOY Controls                     None          A-8           A-l         A-l  &  A-8
  A
Water-Fuel Ratio
Fudl                             DF-2          DF-2           DF-2            DF-2
Stack Effluent
  Flow Rate - Ib/sec
  Temperature - °F
  Water vapor - Vol.  %
  C02 - Vol. * (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 02
ppm @ 1 5% Op & EF
Ib/hr
Carbon Monoxide Emissions

108
108



156 170 160
156 170 160



139 124 132
139 124 132



132
132


  ppm (dry)
  ppm 9 15% 02              248    17.7  99.8  11.7     429     33.0    231      22.2
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 02                9.9   1.0    --    --      26.8   2.4
  Ib/hr

NOTES:
1.  Efficiency factor of no benefit.
2.  Output of engine^not necessarily of test rig.
                                     C-28

    -------
                                       TABLE 12
                                      Facility Kl
                                  Summary of Results
Unit Type
Test Number
Test Date
% Rated Output (0.75 mw)
NO  Controls
  x
Water-Fuel Ratio
Fuel
Stack Effluent
  Flow rate - Ib/sec
            - SCFM
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrggen Oxides Emjss ions
  ppm (dry)
  ppm @ 15% 0,,
             2
  ppm 9 15% 02 (<> EF
  Ib/hr
Carbon Mgnpxide Emjssions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emi ssions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
             Production
     1A                     IB
           2/72 - 4/72
     0                    100
               None

               DF
11,300
   476
     0.5
    19.0
     3.0
     9.0
     9.0
   139
   390
    14.0
    42.0
11,000
   850
     3.0
    16.5
    14.5
    19.3
    19.3
    90
   120
     9.0
    12.0
Notes:
1.  Efficiency factor of no benefit.
                                           C-29

    -------
                                     TABLE  14
                                 Facility  L
                                Summary of Results

    Unit Type    Radial  Industrial  Gas  Turbine  Generating  Set
    Test Number                                 1A    1B
    Test Date
    %  Rated Output (1.0 mw)                      0  100
    NO  Controls                             None   None
       x
    Water-Fuel Ratio
    Fuel                                     DF-2   DF-2
    Stack  Effluent
       Flow rate  - Ib/sec
       Temperature -   F
       Water vapor - Vol. %
       C02  - Vol. %  (dry)
       02 - Vol.  % (dry)                      19.0     16.2
    Visible Emissions - %  opacity            0         0
    N i trogen  Oxj des  Emjssions
       ppm  (dry)                              30         52
       ppm  & 15%  0                            90         69
       ppm  @ 15%  Op  G>  EF(1)                   90         69
       Ib/hr                                 4.6        7.4
     Carbon Monojdde Emissions
       ppm  (dry)                              100        40
       ppm  (3 152  0                            300        54
                  2
       Ib/hr                                 9.4        3.5
     Hydrocarbon  Emissions
       ppm  (dry)                      -      -^10      
    -------
                 TABLE  15

             Facility M

            Summary  of  Results
     Unit Type

     Test Number

     Test Date

     % Rated Output (2.5  mw)

     NO  Controls
       x

     Water-Fuel Ratio

     Fuel

     Stack Effluent

       Flow rate - Ib/sec

       Temperature -  F

       Water vapor - Vol. %

       C02 - Vol. % (dry)

       02 - Vol. % (dry)

     Visible Emissions - % opacity

     Nj trpgen Oxi des Emj ss i ons

       ppm (dry)
Production 1 & 2 Shaft Gas Turbines
       ppm @ 15%
       ppm @ 15$
       Ib/hr
EF
                  9/72-8/73

                    100

                    None




                   Kerosene
                    800
                     16.5%
65


86
82
     C a rbon Mon oxj de Emj ss i ons

       ppm (dry)                           95

       ppm @ 15% 0    ...                 126

       Ib/hr


     Hydrocarbon Emi s si ons

       ppm (dry)


       ppm (3 15% 02

       Ib/hr
Notes:
1.  Data represents average of tests on 3 units.
O    ,|       I)        II     II    II   II  O  II
                        2(2)


                       9/72-8/73

                         100

                         None




                       Natural Gas
                          800
                          17.0%
59


89
85
                                                  45

                                                  68
                     C-31

    -------
                                     TABLE 16
                                     Facility N
                                 Summary of Results
        15% 02 0 EF
                                 0.93
Unit Type
Test Njmber                      1A
Test Date                      5/3/75
% Rated Output (2.5 mw)          0
NO  Controls
Water-Fuel Ratio
Fuel                             N.G.
Stack Effluent
  Flow rate - Ib/sec
  Temperature -  F
  Water vapro - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)
Visible Emissions - % opacity
Ni trogeri Oxi des Emi ss ions
  ppm (dry)
  ppm 0 15% 02
  ppm
  Ib/hr
Carbon Monoxide Emjssions
  ppm (dry)
  ppm @ 15% 02                 236
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 0_                 155
  Ib/hr
                                           Production Prototype Gas Turbine
34
34
                                          IB
                                        5/3/75
                                        100
                                          N.G.
107
103
                                         34
                                         11
                   2A
                 5/3/75
                   0
                                                     None
                    2B
                  5/3/75
                  100
                  (1)
                   Kero
                                                   0.4
65
65
                  95
                  37
                    Kero
          2.36     1.18     2.92
                            3.0
156
149
                   30
                   3A       38
                 5/3/75   5/5/75
                   0      100
                    DF-2    DF-2
                              1.22    2.99
                                      3.8
78     160
78     153
                  108
                             31
                   65
                           10
Notes:
1.  Air atomizing combustor used for smoke reduction.
                                           C-32

    -------
                                    TABLE 17
                                   Facility 01
                               Summary of Results
Unit Type
Test Number
Test Date
% Rated Output (2.5
NO  Controls
  x
Ambient Conditions -
  Temp. °F
  Pressure - psia
  R.H. - %
Fuel
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (Actual)
  02 - Vol. % (dry)
Visible Emissions - % opacity
N1trpgen Qxjdes Emis sions
  ppm (dry)
  ppm & 15% 02
  ppm @ 15% 02 @ EF
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
 Recuperated 2 Shaft  R & D Gas Turbine
1
6/74
100
A-l
66
14.7
76
N.G.
2.13
17.5
0
93
159
118
2
7/74
100
A-l
69
14.7
59
N.G.
1.97
17.5
0
69
119
88
3

100
A-l



DF-2
2.6
17.1
0
196
302
224
4

W2)
A-l



Kerosene
2.5
17.2
0
186
294
218
5
7/74
100
A-l
69
14.7
59
NAPTHA
2.45
17.7
0
173
315
233
17
 3.1
39
 4
19.5
32
20.4
                   2.0
Notes:
1.  Power output assumed same as for simple cycle version.
2.  Assumed, based on fuel flow.
                                     C-33

    -------
                                    TABLE  18
                                Facility  01
                               Summary of Results

   Unit Type-'   Recuperated 2 Shaft R&D Gas Turbine
   Test Number                                          1
   Test Date                                           6/74
   % Rated Output  (2.5 mwr1'                           0
   ML  Controls                                         Al
     x
   Ambient Temp.  - >°F                                   66
           Pressure - PSIA                              14.7
   Fuel    Relative Humidity -  %                        76
                                                        M * U •
   Stack Effluent
     Flow rate - Ib/sec
     Temperature -   F
     Water vapor - Vol. %
     C02 - Vol. %  (dry)                                 0.99
     02 - Vol. %  (dry)
   Visible Emissions -  %  opacity
   Nitrogen Oxides  Emjss jons
     ppm  (dry)
     ppm 0 15% 09                                       31
     ppm 0 15% 0/0  EF                                 31
     Ib/hr      z
   Carbon Monoxide Emjssjons
     ppm  (dry)
     ppm  §  15% 0                                        4,695
                2
     Ib/hr
   Hydrocarbon Emjssj ons
     ppm  (dry)
     ppm  0  15% 02                                      6,011
     Ib/hr
Notes:   1.   Power output assumed same as for simple cycle version.
                                        C-34

    -------
                                     TABLE 19

                                 Facility  PI

                                Summary of Results




    Unit Type :   Production  Gas Turbine


    Test Number                                   1              2A     2B


    Test Date

    % Rated Output (2.5 mw)                       100               0     100

    NO  Controls                                 None                 None*
      x

    Water-Fuel Ratio

    Fuel                                         N.G.                DF-1


    Stack Effluent

      Flow rate - Ib/sec

      Temperature -  F


      Water vapor - Vol.  %

      C02 - Vol. % (dry)

      02 - Vol. % (dry)                          15.69          --      15.79


    Visible Emissions - % opacity                 1                       7

    Nitrogen 0x1des Emi ssi ons

      ppm (dry)                                   88            --     109


      ppm G>15% 02                                99            66     ]^|
      ii   	& fl EF
      Ib/hr                                       16.7          --     20.7


    Carbon Mgn oxj de Emi s s i o n s

      ppm (dry)                                   10            —     50

      ppm @ 15% 0                                 11           376     57
                 2
      Ib/hr                                      0.30           —     1.48


    Hydrocarbon Emjssjpns

      ppm (dry)

      ppm 9 15% 02

      Ib/hr

*Combustor and fuel nozzle modified for smoke and'"part1culate emissions control,
                                           C-35

    -------
                                     TABLE  20
                                 Facility   P2
                                Summary  of  Results


    Unit Type  :   Prediction Gas Turbine
    Test Number
    Te>t Date
    % Rated  Output (2.5 mw)
    NO  Controls
      x
    Water-Fuel Ratio
    Fuel
    Stack Effluent
      Flow rate - Ib/sec
       ''   "      scfm
      Temperature -  F
      Water vapor - Vol.  %
      C02 - Vol.  % (dry)
      02 - Vol. % (dry)
    Visible Emissions - % opacity
    N i trogen Oxj des Em j s s j pns
      ppm (dry)
      ppm @ 15% 0       ,?}
      n   "  »  "^  @ £f(t)
      lb/hr
    CarbojiMonoxide Emissions
      ppm (dry)
      ppm (a 15% 0
      lb/hr
    Hy drgcarbon Emi s si on s
      ppm (dry)
      ppm @ 15% 02
      lb/hr
Notes:
1.  Engine Modification Made to Reduce Particulates
2.  Design heat rate used for efficiency factor calculations.
                                        C-36
1A IB
May 1972
0 100
None
DF
39,100 39,400
380 640
1.10 2.26
19.5 17.0
18 78
72 117
72 111
23.6
—
__
8.0 4.0
32.0 6.0
ft 2B
May 1972
0 100
None * '
DF
39,100 39,400
380 640
1.00 2.53
19.6 17.5
8.5 83
36.4 142
36.4 134
26.2
42 35
180 60
7.0 4.0
30 6.9

    -------
                  TEST  RIG
Facility No.
Unit Type :
Test Number
Test Date
%  Rated  Output(2.5 mw)
     TABLE  21
 Facility   Q &  R
Summary of Results
                   Q
          Production Combustor
   1A      IB     2A    28    2C

       100              100
     R & D Combustor

2D       1A       IB

         0        100
NO Controls B-2 B~}
vft Ratio 0 1.27 0 0.3 0.6 0.9
Water-Fuel Ratio DF-2 DF-1
Fuel
Stack Effluent
Flow rate - Ib/sec
Temperature - °F
Water vapor - Vol. %
C02 - Vol. % (dry)
02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 00 , % 113 68 154 83 44 32
:" 	 2@EF(1) 112 64 146 79 42 30
Ib/hr
Carbon Monoxide Emissions
ppm (dry) •
ppm @ 15% 0 45 54 33 30 36 38
2
Ib/hr
Hydrocarbon Emissions
ppm (dry)
A-4 & A-l
DF-1









140 50
140 47

200 4


       ppm @ 15% 02
       Ib/hr
Note:  ].  Design heat rate of production engine used for efficiency calculations
                                           C-37

    -------
                                    TABLE  22
                                Facility   S
                               Summary of  Results


   Unit Type :
   Test Number                           1A         1B              2
   Test Date
   %  Rated  Output (6-° mw-base)         100        125             125
   NO Controls                               A'9                  None
      x
   Water-Fuel Ratio
   Fuel                                   Distillate Oil          Natural Gas
   Stack  Effluent
      Flow rate  -  Ib/sec                 -   98,500                98,500
      Temperature  -   F
      Water vapor  - Vol.  %
      C02  -  Vol.  %  (dry)
      02  - Vol.  %  (dry)                  16.5      16.0                16.0
    Visible Emissions - % opacity
    Nj trpJg" Oxi des  Em i s si pns
      ppm (dry)                           66       75                  30
      ppm @ 15%  05     m               '88       90                  36
       	* @EP  '               88       90                  36
      Ib/hr                                        57                  23
    Carbon Mgngxjde Emissions
      ppm (dry)
      ppm (3 }$% 0
                 2
      Ib/hr
    Hydrpcarbon Emi s s j on s
      ppm (dry)
      ppm @ ]5% 02
      Ib/hr
Notes:
1.  Efficiency factor of  no benefit

                                       C-38

    -------
                                 TABLE 23
                             Facility  T-l
                            Summary of Results

Unit Type :   Pipeline Pumper
Test Number                          1                    *
Test Date
% Rated Output (10.3 mw)            100                  100
N0x Controls                        None                 None
Water-Fuel Ratio
Fuel                                OF                   N.G.
Stack Effluent
  Flow rate - Ib/sec              125,400                125,400
  Temperature - °F                    775                    735
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)                   17.7                  17.7
Visible Emissions - % opacity
Nj trogen Oxides Emj s s i ons
  ppm (dry)                           30                    15
  ppm G> 15% 09     ...                54                    27
  11   "  "  "*  @EF(1'                54                    27
  Ib/hr                              30                    15
Carbon Monoxi de Emj ss i ons
  ppm (dry)                          90                    50
  ppm 0 152 0                       162                    90
             2
  Ib/hr BTU                         0.38                   0.21
Hydrpcarbon Emi ss i ons
  ppm (dry)
  ppm @ 15% 02
  Ib/hr
Notes:
1.  Efficiency factor of no benefit.
                                    C-39

    -------
                                 TABLE 24
                             Facility T2
                            Summary of Results

Unit Type :   Prod- .tion Pump Driver
Test Number                   1A                IB            2A         2B
Test Date                              10/72                    10/72
% Rated Output  (10.3 mw)      0                 107           17         102
NO  Controls                      None                            None
  x
Water-Fuel Ratio
Fuel                                   DF-1                       N.G.
Stack Effluent
  Flow rate - Ib/sec          30-3               T67-1          87-2      161-1
  Temperature -  F
  Water vapor - Vol.  %
  C02 - Vol. %  (dry)
  02 - Vol. % (dry)           18.0                17.6          18.5       17.7
Visible Emissions - % opacity
N i trgc|en Oxi des Emi s s i ons
ppm (dry)
ppm @ 15% Q
" " " " 0 EF
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm 0 15* 0
Ib/hr
i.ydrocarbon Emissions
ppm >(xta!j0& (wet)
ppm 0 15% 02
Ib/hr
13.4
26.8
26.8
2.68
463
926
56.3
287
574
17.9
85.4
151
146
79.35
35.4
62.5
20.1
1.03
1.82
0.30
12.1
5.04
5.04
6.51
197
473
65.0
31.9
76.6
6.26
40.8
74.2
71.5
36.2
85.3
155
45.7
2.3
4.2
0.75
                                      C-40

    -------
                                     TABLE  25
                                 Facility  Ul
                                Summary of Results



    Unit Type:   Field Test, Simple  Cycle "Peaking"  Power Plant
Test Number
Test Date
% Rated Output (15.1 peak
NO Controls^
Water-Fuel Ratio
Fuel
Stack Effluent
Flow rate - Ib/sec
Temperature - F
Water vapor - Vol. %•
C02 - Vol. % (dry)
0? - Vol. % (dry)^'
1A
5/72
MW)86 100
B-l
0.0




1.8 --
17.6 "17.3
IB 1C ID
5/72 5/72 5/72
86 100 86 100 86 100
B-l B-l B-l
0.5 .75 1.0
N G \




1.8 — 1.8 -- 1.8 —
17.6 17.3 17.6 17.3 17.6 17.3
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry)
V 1 IS" »z w/EFl"
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm @ 15$ 0
2
Ib/hr
Hydrocarbon Emissions
pprr ^o.)
ppm @ 15% 02
Ib/hr
68 82
120 149
120 148

28 46
49 84

44 26
78 47

34 44 24 40 15 29
60 80 42 73 26 53
60 79 42 72 26 53

109 87 137 277 214 233
192 158 242 504 378 424

7* 4F 164 111 Of M*
124 82 184 204 228 215

Notes:
1.  G.T. Equipped with smokeless combustor
2.  Peak load Op percentage estimated
3.  Efficiency factor of no benefit at base load.

                                          C-41

    -------
                                    TABLE   26
                                Facility  U2
                               Summary of Results


    Unit Type-'   Field Test, Simple Cycle "Peaking" Power Plant
    Test Number                 1A   IB      2A    2B        :
    Test Date                                 1/71 - 2/71
    % Rated  Output (15.1 peak mw) 86  116      86   116       (
NO controls y 	 °-' 	
X
Water-Fuel Ratio 0.0 0.0 0.5 0.5
Stack Effluent
Flow rate - Ib/sec
Temperature - °F
Water vapor - Vol . %
C02 - Vol. % (dry) 1.8 — 1.8 —
02 - Vol. % (dry)O) 17.6 17.3 17.6 17.3
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry) 59 66 22 26
pjpm $ 1,5% p2 /EF(2) 104 ]20 39 47
lb/hr
Carbon Monoxide Emissions
ppm (dry) 53 58 286 290
ppm (3 15% 0 94 105 505 527
2
lb/hr
ilyjirocarbon Emissions
ppm (dry) 30 20 124 124
ppm @ 15% 00 53 36 219 225
	 7
0.75 0.75 1.0 1.0




1.8 — 1.8 —
17.6 17.3 17.6 17.3

16 19 9 15
28 35 16 27
28 35 16 27

413 344 567 354
729 625 1001 644

208 160 400 ?fi
    -------
                                 TABLE  27
                             Facility  U3
                            Summary of Results


Unit Type:   Field Test, Simple Cycle Peaking Power Plant
Test Number 1
Test Date 8/69
X Rated Output (15.1 peak mw) 100
NO Controls None
X
Water-Fuel Ratio N/A
Fuel NG
Stack Effluent
Flow rate - Ib/sec 783,000
Temperature - F 720
Water vapor - Vol. %
C02 - Vol. % (dry) 1.9
02 - Vol. % (dry) 17.6
Visible Emissions - % opacity
Nitrocjen Oxides Emissions
ppm (dry) 35
ppm (a 15% 0, 62
	 * w/EF 62
Ib/hr
Carbon Monoxide Emissions
ppm (dry) ^.]Q
ppm @ 15% 0 
    -------
                                 Facility  VI
                                Summary of Results


    Unit Type :   Si..,pie Cycle "Peaking" Power Plant
Test Number ^ '
Test Date
% Rated Output07.2 peak mw)
NO Controls '
X
Mater- Fuel Ratio
Fuel
Stack Effluent
Flow rate - Ib/sec
- DSCFM
1A
2/72
no
B-l
0
N.G.

mm ^
2A
2/72
106
B-l
0
DF-2

213 300(3
C 1 O ) OWU
IB
2/72
no
B-l
=0.5^
N.G.

* _ —
2B
2/72
106
B-l
0.52
DF-2

"• •*-
      Temperature - °F

      Water vapor - Vol. %
      Ib/hr
See Notes on attached sheet
C02 - Vol. % (dry) 2.82
02 - Vol. % (dry) 15.3
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry) 88
ppm (3 15% 09 (6) 92
" " " "' w/EF 92
Ib/hr
Carbon Monoxide Emissions
ppm (dry) <10l
ppm (3 15% 0 <10
2
Ib/hr
Hydrocarbon Emissions
ppm (dry) 11.5^
ppm (a 15% 00 12.1
3.53 2.98
15.1 15.1

110 38
113 38
113 38
.10
<10
7.4(2) 8.9
7.6 9.0
3.4
14.9

54
53
53
-_
_ —
4.2
4.1
                                        C-44

    -------
                                  Table 28
                                  Facility VI


 Notes:
1.   Average of 3 runs
2.   Data available for only 2 runs
3.   "     "         "   "   1 run.
4.   Gas flow not recorded, but water flow comparable to Test 4.
5.   Air atomization added to reduce plume opacity
6.   Efficiency correction of no benefit.
                                       C-45

    -------
                                 Facility  VI
                             Summary  of Results
Unit Type:  Field test,  Simple Cycle  Peaking  Power  Plant
Test Number                             1C                  2C
Test Date                               2/72                 2/72
% Rated Output (17.2 mw  peak)           26.9                 29.7
NO  Control s^                         None                 None
  /V
Water-Fuel Ratio
Fuel                                    N.G.                 DF-2
Stack Effluent
  Flow Rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
  C02 - Vol. % (dry)                    1.18                1.38
  02 - Vol. % (dry)                     18.1                 17.78
Visible Emissions - % opacity
Nitrogen Oxides Emissions
  ppm (dry)                              22                   31
  ppm @ 15% 02                           45                   55
  ppm @ 15% 02 w/EF(2)                   45                   55
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)                            150^               220
  ppm @ 15% 02                         310                  398
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)                            20.3^              11.3
  ppm @ 15% 02                         42.0                  20.4
  Ib/hr
SEE NOTES ON ATTACHED SHEET
                                    C-46

    -------
                                 Table 29
                                 Facility VI

Notes:
1.  Average of 3 runs
2.     "    "  2 runs.
3.  Data available for only 2 runs
4.   "     "       "    "   1 run.
5.  Air Atomization added to reduce plume opacity
6.  Efficiency correction of no benefit.
                                   C-47

    -------
                                     TABLE  30
                                 Facility   V2
                                Summary  of  Results


    Unit  Type  •'   FieH Test, Simple Cycle "Peaking" Power Plant
    Test  Number
    Test  Date
    %  Rated  Output(17.2 mw, Peak)
                (4)
    NO Controls
      x
    Water-Fuel Ratio
    Fuel
    Stack Effluent
      Flow rate  - Ib/sec
      Temperature -  °F                  972          972         1050           625
      Water  vapor -  Vol.  %
      C02 -  Vol. % (dry)
,01
1/73
94.2
B-l
0.43
DF-2
2A(2)
1/73
93.8
B-l
0.43
DF-2
2B(3)
1/73
107.6
B-l
0.42
DF-2
2C(3)
1/73
29.1
B-l
0
DF-2
02 - Vol. % (dry)
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 0? (5)
ii H ii lit w/EF
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm 9 15% 0
2
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% 00
15.5
10
51.8
56.4
56.4
59.3
28.2
30.9

0.77
0.84
15.2
10
49.1
50.4
50.4
53.4
33.3
34.2

0.50
0.52
14.4
—
61
56
56
20
18.8

2.5
2.2
17.5
—
32
55
55
235
403

9.5
16.3
      Ib/hr
See notes on attached sheet.
                                         C-48

    -------
                               Table 30
                               Facility V2
Notes:
1.  Average of 3 runs, unit GT-2A,  49 point traverse
2.  Average of 2 runs, unit 6T-2B,  "   "
3.  Single point runs,  "   GT-2B, (one run each, no traverse)
4.  Air atomization added to reduce plume opacity
5.  Efficiency correction of no benefit.
                                    C-49

    -------
                                    TABLE  31
                                Facility  V3
                               Summary of Results

   Unit Type:  Field Test, Simple Cycle Peaking Power Plant
   Test Number ^                  1             2             3              4
   Test Date
   % Rated Output (17.5 peak mw)  100           100           100            100
   NO  Controls                   None            A            B-l           C-l
   Water-Fuel Ratio               N/A           N/A           0.56           0.56
   Fuel                           DF-2         DF-2          DF-2           DF-2
   Stack  Effluent
     Flow rate -  Ib/sec
     Temperature  -   F
     Water vapor - Vol. %
     C02  - Vol. %  (dry)
     02 - Vol. %  (dry)           15.5         15.5          15.5            15.5
   Visible Emissions - %  opacity
   Nitrogen  OxjdesEmissions
ppm
ppm
it
(dry)
8 15%
H n
°2
II £
w/EF
(2)
150
164
164
100
109
109
75
82
82
57
62
62
      Ib/hr
    Carbon Mgngxide Emissions
      ppm (dry)                   -                -           500
      ppm @ 152 0                 —                ~           547
      Ib/hr
    Hy d r oc a r b o n Em i s s j on s
      ppm (dry)
      ppm @ 15% 02
M *   lb/hr
Notes:
 1) Number of tests used to provide this data base is unknown.
 2)  Efficiency correction of no benefit.
                                           C-50

    -------
                                     TABLE   32
                                 Facility   V3
                                Summary of  Results
    Unit Type:   Field test,  simple  cycle  "peaking" power plant
    Test Number'1'             5                  678
    Test Date
    % Rated Output (17.5 peak  mw)100
    NO  Controls
      x
    Water-Fuel  Ratio
    Fuel
    Stack Effluent
      Flow rate - Ib/sec
      Temperature -  F
      Water vapor -Vol. %
      C02 - Vol. % (dry)
      02 - Vol. % (dry)       15.5             15.5              15.5            15.5
    Visible Emissions - % opacity
    Njtrogen_0x1des Emi ssions
iw)100
None
N/A
NG
100
A
N/A
NG
100
B-l
UNK
NG
100
C-l
UNK
NG
ppm (dry) 85 68
ppm @ 15% 09 ,,; 93 74
" 	 d w/EFu; 93 74
Ib/hr
Carbon Monoxide Emissions
ppm (dry) 500
ppm @ 15% 0 547
2
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% Op
Ib/hr
39 30
43 33
43 33
100
110
Notes:
(1) No of tests used to provide data base 1s unknown.
(2) Efficiency correction of no benefit.
                                           C-51

    -------
                                 TABLE 33
                             Facility  W
                            Summary of Results

Unit Type:  Fie'j test, Simple Cycle "Peaking" Power Plant
Test Number                          ^
Test Date                            "/72                11/72
% Rated Output(2K3 mw max.  peak)     23                   85
NO  Controls                         D"]                  D"1
  x
Water-Fuel Ratio                      °                   °-6
Fuel                                 JP'5                  JP'5
Stack EffluentD$CFM                  _.                   256>QOO
  Flow rate - Ib/sec
  Temperature - °F                   --                    573
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)                  18.5                   17.8
Visible Emissions - % opacity         --                    ^-^
N i trogen  Ox i des Emj s s i ons
  ppm (dry)                          20                    31.1
  ppm (3 15% 00                       48                    62.2
          »  »2  w/EF                 48                    56.7
       ••
      Ib/hr                              --                    61-2
    Carbon Monoxide Emissions
      ppm (dry)                          222                   43.7
      ppm 9 15% 0                        533                   74
                 2
      Ib/hr
    nydrgcarppn Emissjons
      ppm (dry)                          ~                     2.9
      ppm (<> 15X 02                       ~                     5.4
      Ib/hr
Notes: (1) 1 single point test (run 5)
       (2) average of 3 runs of 44 traverse points each.
                                      C-52

    -------
                                 TABLE   34
                             Facility   X
                            Summary of Results

Unit Type:  Field Test, Simple Cycle "Peaking" Power Plant
Test Number                          1A                     IB         EST.
Test Date                           12/74                  12/74
% Rated Output(27.6 mw peak)          70                     70         100
NO  Controls                                       N°ne
  x
Water-Fuel Ratio
Fuel                                DF-2            M(1)                M(1
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor -Vol. %
                                    16.4           16.4                15.5
      C02 -  Vol.  % (dry)
      02 - Vol.  % (dry)
    Visible  Emissions - % opacity
                      - Von Brana        -              98-99
    Nj trgjjen Qxj des Emj ss igns
      ppm (dry)                          129                34                55
      ppm @  15%  0,                      168                44                60
      ppm (3  15% #?, w/EF                141                37                50
      Ib/hr        c
    Carbon Monoxide Emissions
      ppm (dry)                          22                48                15
      ppm (3  15%  0                       28                62                16
                 2
      Ib/hr
    Hydrocarbon  Emissions^
      ppm (dry)
      ppm @  15%  02
      Ib/hr
Notes: (1)   Methanol  fuel  heat  of  combustion was 8550  Btu/lh.
       (2)   Fuel system sized for DF-2, which has about twice the heat of
            combustion, so full  power output could not be achieved.
                                     C-53

    -------
                                 TABLE 35
                             Facility  Y
                            Summary of Results


Unit Type:  Fiel.. Test, Peaking Unit
Test Number                        1A                IB              2
Test Date                          	12/71 - 2/72	
% Rated Output(32.8 mw peak)       21                99               99
NO  Controls                       B-l               B-l              B-l
  /\
Water-Fuel Ratio                   °                 °                °'7
Fuel                            	DF-2	
Stack Effluent
  Flow rate - Ib/sec
  Temperature - °F
  Water vapor - Vol. %
C02 - Vol. % (dry)
02 - Vol. % (dry)
Visible Emissions - |mftg|C^
(AS
Nitrogen Oxides Emissions
ppm (dry)
pom @ 15%.,00
II II II II £ W/£F
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm @ 15% 0
lb/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% 02
lb/hr
1.0
17.0
Sit 34 %
TM D-2156)
44
66
66
•»"•
29
44

<0.96
<1.4
--
3.0
14.1
4^8
217
189
163
"•"•
36
31

£: .96
< .84
—
3.0
14.1
58
50
43
• MM
11
10

0.80
0.70
--
                                    C-54

    -------
                                 TABLE 36
                             Facility  Zl
                            Summary of Results


Unit Type:   Combustor Rig  Tests
Test Number                       ™         IB              2A        2B
Test Date
% Rated Output ("^w)          122          122            94         94
NO  Controls                    None          A-l            A-2<2>    A-2<3>
  x
Water-Fuel Ratio
Fuel                            DF-2          DF-2          N.G.         N.G.

Stack Effluent
  Flow rate - Ib/sec
  Temperature -   F
  Water vapor -Vol. %
                    )
                    (4)
C02 - Vol. % (dry)
   02  -  Vol.  %  (dry)
 Visible Emissions  -  % opacity
 Nitrggen 0x1des  Emjssigns
   ppm (dry)                    240             111           87
   ppm 9 15%  0_ (4)
    ii   n n   ii 2
   Ib/hr
 Carbon Monpxj de Emj ss ions
   ppm (dry)
   ppm @ 15% 0
              2
   Ib/hr
 Hy d r pc a r bp n Em j s s j o n s
   ppm (dry)
   ppm @ 15% 02
   Ib/hr
                                      C-55

    -------
                               TABLE 36
                               Facility Zl

Notes:

(1)  Power output not stated, full  load assumed equivalent to a .017
     fuel/air ratio.
 2   Primary airflow 35$ of total  airflow.
 3   Primary airflow 45% of total  airflow.
 4   Percent (L 1n exhaust gas not noted in report.
                                      C-56

    -------
   Test Number
   Test Date
   % Rated Output



ustor R1g
1A/2A
TOO
None


TABLE 37
Facility Z2
Summary of Results
Test, Peaking Unit
IB 1C ID 2B 2C
100 100 100 100 100
A-3 A-la&A-3 A-lb B A-l & B
1.0 1.0
	 DF-2 	
   NO  Controls
     A
   Water-Fuel Ratio
   Fuel
   Stack Effluent
     Flow rate - Ib/sec
     Temperature -  F
     Water vapor - Vol . %
     C02 - Vol. %  (dry)
     02 - Vol. % (dry) 14.0    14.0      14.0        14.0      14.0        14.0
   Visible Emissions  - % opacity
     trogen  Oxj des  Emj ssi ons
ppm (dry)
ppm @ 15% 0
ii ii ii ii £• W/hr
lb/hr
Carbon Monoxide Emi
ppm (dry)
ppm @15«0
2
lb/hr
202 140 115 96 40
173 120 99 82 34
149 104 85 71 29
ssions



19
16
14



Hy d roc a rbpn Emj s s i on s
      ppm (dry)
      ppm 9 15* 02
      Ib/hr
Notes:  (1) output, not stated in report,  is  around  28 mw peak  for eftigine.
                                        C-57

    -------
                                    TABLE 38

                                Facility  AA1

                               Summary of Results
Unit Type: Scaled Combustor Rig Test
Test Number
Test Date
% Rated Outpu/tj\ /2)
NC Controls
A
Water-Fuel Ratio
Fuel
Stack Effluent
Flow rate - Ib/sec
Temperature - °F
Water vapor - Vol . %
C02 - Vol. X (dry)
02 - Vol. X (dry)
Visible Emissions - %
1A IB

100 100
None A-5
N.G. N.G.

2 2

4.83 4.8.3
4.83 5.55
16.1 14.3
opacity
2A

100
None
DF-2

2

4.83
5.55
14.8

2B

100
A-5
DF-2

2

4.83
5.55
12.5

3A

100
None
DF-2

1.95


15.0<3>

3B
•
100
D-2
7.0^
DF-2

1.95

—
15.0«:

Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 09 (5\
ii H ii ii c w/EF
Ib/hr
60 44
73 39
63 34
82
79
68
52
37
32
73
73
63
25
25
22
Carbon Monoxide Emissions
ppm (dry)
ppm @ 152 0
2
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15X 02
43 43
53 39

2 0
2.4 0
70
68

6.5
6.3
30
21

0
0
50
50

0
0
200
200

7.5
7.5
      Ib/hr
SEE NOTES ON ATTACHED SHEET.
                                          C-58

    -------
                            TABLE 38
                            Facility AA1
NOTES:
 1   Output not stated in report.
 2  1800°F combustor exit temp,  assumed equivalent to full  load output.
 3  02 content assumed, with no change for steam injection.
 4  Fuel/air ratio assumed to be  .017 at full  load.
 5  Combustor efficiency assumed  equiv. to full  size unit.
                                  C-59

    -------
                                  TABLE  39
                              Facility   AA2
                             Summary  of  Results


 Unit Type :  Scal^J Combustor R1g Test
 Test Number             1A        IB          1C           2A            2B
 Test Date
 % Rated, Autput(!lW4) 100       100         100         100            100
 NO  Controls             None      A-la       A-lb         None           A-lb
   A
 Water-Fuel Ratio
 Fuel                    N.G.      N.G.         N.G.        DF-2           DF-2
 Stack Effluent
now rate - ib/sec 	
Temperature - °F
Water vapor - Vol. %
C02 - Vol. % (dry)
02 - Vol. % (dry/2) 16.1
Visible Emissions - % opacity
- ASTM smoke
Nitrogen Oxides Emissions
ppm (dry) 61
ppm @ 15% 0, .....(3) 75
ii i 	 2; w/EFv ' 55
Ib/hr
Carbon Monoxide Emissions
ppm (dry) 38
ppm (3 15% 0 47
2
Ib/hr
H: Irocarbon Emissions
ppm (dry) 7
ppm @ 15% 00 9


16.1
0

53
65
56


26
32


—
—
	 1 .33 	

16.1
0

49
60
52


21
26


5
6


14.8
41

70
68
59


105
102


6
6


14.1
26

61
59
51


85
82


6
6
Notes: (1) output not stated in report.  (2)  Percent 02  not stated in  report, assumed
same as for unit with EGR with no change for dry  controls. (3)  Combustor efficiency
assumed equivalent to full size combustor.
(4) 1800°F output temp,  assumed equivalent  to full size load  output.

                                      C-60

    -------
                             r ai i i i
                                       on
                           Summary  of Results
                                      (1)
Unit Type    Scaled Combustor Rig  Test
Test Number              1A        IB
Test Date               	
                      »
% Rated Output (~MW)
                                            1C
                              ID
NOV Controls
  A
Water-Fuel Ratio
Fuel
Stack Effluent
       None
DF-2
                       1800
SCG
             (2)
           1742
DF-2  SCG
                      (2)
                            2A
                            A-2
                                                             DF-2
             2082   2112    1950
Temperature - F
Q~ - Vol % (Dry)
inlet Temperature F  —	*  620-770  	
Inlet Pressure - Atm. 	3-6-^	
Fuel/Air Ratio        	.026 -  .028	
Space Velocity
Reference velocity
  FT/SEC             130         92          73     105      74
Combustor Pressure
  PSIA                42         51.1         56.5   44.0    47.2
                                                                   2B
                                                                  •4/74-
                                 2C
                                                                    DF-2
                         2173
                                                                        112
                                                                        43.9
SCG'
                                                                                 2220
Nitrogen Oxides Emissions
  ppm (dry)
  ppm (3 15$ 02          60
  Ib/hr
Carbon Monoxide Emissions
  ppm (dry)             18
  ppm @ 15% 02
  Ib/hr
Hydrocarbon Emissions
  ppm (dry)              8
  ppm @ 15% 02
  Ib/hr
    Notes:  See attached sheet
           160
            80
              10     40
                     14
                20
                18
                                                                        30
                                                                               26
                                     C-61

    -------
                              TABLE 40
                              Facility BA 1

Notes :
(1)  Same test rig used for all  tests, with  6"D
     combustors installed for tests 1A, IB and  2A and 6"D
     catalyst cores installed for tests 1C,  ID, 2B and 2C
(2)  molar composition of synthetic coal  gas  was  =  H5  -  15.42.,
     CO = 11.6% methane - 5. IX,  CO, = 10. IX,  N9 - 5778%.
     Lower heating value = 126 Btufft.  .       £
                               C-62

    -------
Unit Type:   Scaled  Combustor Rig Test
                              Facility  BA 2
                            Summary of Results
                                    0)
Test Number
Test Date
% Rated Output
1A
IB
1C
ID
                                                       IE
IF
   ppm (dry)         1         1
   ppm @ 15% CL
   ppm 9 15% 02 w/EF
   Ib/hr
Carbon Monoxide Emissions
   ppm (dry)        48        10
   ppm @ 15% 02
   Ib/hr
Hydrocarbon Emissions
   ppm (dry)        10         3
   ppm @ 15% Op
   Ib/hr
  NOTES:   See attached  sheet
                                        48
                             15
                             32
                            25
16
                           10
1H
iiu ouri tr u I b
X
Water-Fuel Ratio
Fuel'2'
Inlet Temp. -°C
Fuel /Air Ratio
Space Velocity
(1000 hr."')
REF. Velocity -
Ft/Sec.
P vo ecu VP — A TM
Visible Emission
Nitrogen Oxides
M P
550 365
.024 .022
120 100
— —
s - % opacity
Emissions
	 	 — c 	 	 	 • 	
Doped Dopei
DF-2 JP-4 Jet-A SC6 P DF-2
380 405 400 220 250 360
.027 .022 .027 .313 .032 .025
130 180 140 300 190 130
— — — — — —


                                                                                 450
                                       C-63

    -------
                                 TABLE  41
                              Facility  BA  2
(1)   1"  nominal  diameter test rig  used for all  tests.
(2)   M = Methane,  P =  Propane, Doped P is  propane  with  0.17% N2  as  INH,
     Doped DF-2  contains 0.94% N2  as pyridine,  SCG =  Synthetic coal  gas.
     Conversion  of fuel  N2 is 70-90% compared to about  50% for a diffusion
     flame burner.
                                    C-64

    -------
                                      TABLE  42
                                   Facility  CA
                                  Summary of  Results
                                           (1)
      Unit Type :    Aircraft Engine Test Rigx
      Test Number                 1A           IB             1C                ID
      Test Date
      %  Rated  Output (^mw)      100         100            100             100
      NO Controls                 None       A-13           A-2             A-14
        x
      Water-Fuel  Ratio
      Fuel                        	Jet " A	
      Stack  Effluent
        Flow rate  - Ib/sec
        Temperature -   F
        Water vapor - Vol.  %
        C02  - Vol.  %  (dry)
        02  - Vol.  % (dry)          13.6        13.6           13.6             13.6
      Visible Emissions - %{nQ^cJtv.   #          i             ~6                14
      Nitrogen_0xides  Emissions
        ppm (dry)                    440       190            288               173
        ppm (3 15% 09   .__(3T        357       154            234               140
        	2 W/EF           248       107            162               97
        Ib/hr
      Carbon Mgnoxide Emjssions
        ppm (dry)
        ppm @ 15% 0
        Ib/hr
      Hydrocarbon Emissions
        ppm (dry)
        ppm @ 15% 02
        Ib/hr
<)TES:  See  attached sheet
                                           C-65

    -------
                                 TABLE  42

                                 Facility CA

Notes:
(1)  Test rig simulates engine combustor  inlet conditions  except for
     pressure (6.8 atmospheres versus 21.7 produced in engine).
     Correction factors applied to compensate for this difference.

(2)  Engine produces 44,300 Ib thrust at takeoff (about 30  mw @  1.1/2)

(3}  Based on SFC of .349 #Fuel/#thrust and an estimated 1.1  Ib  thrust
     per horsepower.
                                    C-66

    -------
                                  TABLE  43
                                 Facility DA
                             Summary of  Results

Unit Type:   Advanced Aircraft Engine Combustor Test     ^  '
Test Number               1        2A       2B       3A       3B        4A       4B
Test Date
% Rated Output (:.' mw)   100      69      100       69      100        69      100
NO  Controls             None        None              A-4               A-5
Water- Fuel Ratio
Fuel                     ----------------------- Jet-A ----------------------------
Stack Effluent
   Flow Rate - Ib/sec
   Temperature - °F
   Water vapor - Vol . %
   C02 - Vol. % (dry)
02 - Vol . % (dry)
Visible Emissions -
SAE Smoke #
Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 02
ppm @ 15% 02 w/ EF
Ib/hr
Notes:
13
11.4
(2)
529
397
260


14.5 13 14.5 13 14.5
12 <15 
    -------
                                      TABLE  44
                                      Facility EA
                                  Summary of  Results

Test Type:                                               Field Test
Unit Type:                      •                  Production Simple  Cycle
Test Number:                             1A     IB      2A     2B       3A       3B
Test Date:                                                  7/73
% Rated Output (26 MW base)               20     85    .  20     85       20       85
NO  Controls                '                 None            None            None
  A                      •
Water-Fuel Ratio
Fuel                                         N.G.            DF-2(1'         Heavy Dist.{2'
Stack Effluent
  Flow rate - Ib/sec
  Temperature'- °F                             850           850              850
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. % (dry)                      18.8   16.8    18.7    16.6     18.6    17.2
Visible Emissions - Smoke SBT (0-2156)     0       0       4.7     5.5      4.2     5.1
Nitrogen Oxides Emissions                                                  '•-
  ppm (dry)               ^               52     72      48     137       28      127
  ppm 0 15% 02            •              142     103     125     187       70   .   201
  ppm @ 15% 02 0 E.F.     '                      89            161              173
  Ib/hr                                                             •
Carbon Monoxide Einjssjpns
  ppm (dry)                              20       3      15       6       24        7.5
  ppm @ 15% 02                           55       4.3    39       8.2    60       11.8
  Ib/hr
Hydrocarbpn Emi ss i ons
  ppm (dry)                                0.65   0.25    0.5    0.5     0.56    0.25
  ppm @ 15% 02                             1.8    0.36    1.3    0.68    1.4     0,.39
  Ib/hr

NOTES:
(1)  Sulfur content of DF-2 ranged from 0.024 to 0.0255 percent and averaged 0.136  percen
(2)  Sulfur content of heavy distillate ranged from 0.17 to 0.24 percent and averaged 0.2
     percent.
                                         C-68  '

    -------
                                       TABLE  45
                                   Facility   FA
                                 Summary of  Rosults(1)
     Test type -  Field
     "n1t Type-  Production simple  cycle
     Test Number
                                              1A            IB            1C
     Test Date
                                           1/74 - 8/74 	
     * Rated Output (51.7 mw base)            0           lftn
                                                         I00           ]00
    NO  Controls
      *                                    	B  1  	
    Water-Fuel Ratio                         n  "   -
    c  ,                                                   °             !-12
    Fuel
                                               	uv-c
 Stack  Effluent
   Flow rate - lb/sec
                                                      ouu
   Temperature - °F
                                         —  '      .   900
   Water vapor -  Vol.  %
   C02 - Vol.  % (dry)

   V  'OK*  (dry)

 Visible Emissions - %  opacity          12            8

^f^^^Oxj^^Emjssjon^
  Ppm (dry)
                                       38          230               43
        15X 0
   lb/hr
Note: 1) Data corrected to 150 conditions.
                                                                     16.62
     "   "  "   "2 w/EF                   172           315
    lb/hr                               172           274              |f
                                                     809
                                        55           22
                                                                     80
                                       250           30
                                                     d°      .
                                                    47  2
                                                      '
                                    C-69

    -------
                                 TABLE  46
                             Facility  GA
                            Summary of Results
Test type - Field
Unit Type - Production  Simple  Cycle
Test Number                            1                      2
Test Date                            12/72                10/72 and  12/72
% Rat--' Output (52.9 mw)^            100                     100
NO  Controls                      '               None
  A                '
Water-Fuel Ratio
f.uel                                 DF-2                     Crude
Stack Effluent
  Flow rate - Ib/sec
  Temperature -  F
  Water vapor - Vol. %
  C02 - Vol. % (dry)
  02 - Vol. X (dry)                  16.0                       16.0
Visible  Emissions  -  %  opacity
N1 trogen  Oxides  Emi s s j pns
  ppm  (dry)                           128                      HO
  ppm  0  15% 0.                        154                      168
    "   "   "  "2  w/EF                  135                      147
  Ib/hr
 Carbon Monoxide  Emjssions
  ppm  (dry)
  ppm  @  1525 0
   Ib/hr
 Hydrocarbon  Emi ss i ons
  ppm  (dry)
   ppm
   Ib/hr
 Note:   (1)  A  firing  temp, of 1730°F was assumed equivalent, to full load.
                                    C-70

    -------
                                  inouc <\l
                                Facility HA  1
                             Summary of Results
Test Type - Field
Unit Type - Production Simple Cycle
Test Number
Test Date
% Rated Output
(60.4 mw base)
NO Controls
A
Water- Fuel Ratio
Fuel
Stack Effluent
Flow rate - Ib/sec
Temperature - °F
Water vapor - Vol . %
C02 - Vol. % (dry)
02 - Vol. % (dry)
1A
98.3
0

512



15.60
IB 1C ID IE IF
99.5 101.1 102.3 102.8 103.5
Bi _ -- -- _-
0.23 0.45 0.66 0.87 1.07
__„ 	 	 	 DF-? 	

512 512 511 508 506



15.50 15.40 15.30 15.20 15.10
Visible Emissions - % opacity
Nitrogen Oxides Emissions
ppm (dry)
ppm @ 15% 02
ppm & 15% 02 w/ EF
Ib/hr
Carbon Monoxide Emissions
ppm (dry)
ppm & 15% Op
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% Op
Ib/hr
147
163
130
—
0
0
0
0
0
0
98.8 67.4 44.7 32.5 25.5
108 72.3 47.1 33.6 25.9
85.8 57.4 37.4 26.7 20.6
— — — — —
00 0 4.0 9.0
.0 0 0 4.1 9.2
00 0 	
00000
00000
00000
                                         C-71

    -------
                                 TABLE  48
                             Facility HA 2
                            Summary of Results' '
Test type - Field
Unit Type - Production Simple  Cycle
Test Number           ™&       1B         1C             1D           1E
Test Date
% Rate-  Output
 (61.5 mw base)
NO  Controls
Water-Fuel Ratio
82.9
0
84.6
0.27
86.2
	 o-i 	
0.52
	 — NG -
87.8
0.73
89.4
0.98
Stack Effluent
   Flow  rate - Ib/sec   442      451        453            452          452
   Temperature -   F
   Water vapor - Vol. %
   C02 - Vol. %  (dry)
   02 -  Vol. % (dry)   15.67     15.40      15.25         15.10         14.90
Visible Emissions - % opacity
Njtrogen  Oxides  Emissions
   ppm  (dry)            99.0      60.7        37.4          24.6           13.4
        15% 0,        m        65.0        39.0          25.0           13.2
                  EF
   Ib/hr
ppm @ 15% 0         IN         6b-u         Jy-u           ">u           'X.
pl?m h ',?* ,72..., cc    87.8      51.2         30.7           19.7           10.4
 Carbon Monoxide Emissions
ppm (dry)
ppm @ 15% 0
2
Ib/hr
Hydrocarbon Emissions
ppm (dry)
ppm @ 15% Op
2.14
2.41

--

2.44
2.75
3.00
3.21

--

IvOfi
1.14
3.00
3.13

--

'1.06
1.11
3.00
3.05

•••»
•
1.07
1.09
10.00
9.84



2.15
2.11
         (1)  Gas turtxine inlet air ai'lOO*  relative  humidity  ("air  saturated  using
             evaporative coolers).
         (2)  Average of 7 tests.

    -------
                                      TABLE  49
                                  Facility J A
                                 Summary of Results
    Test type:  F1dld
    Unit Type :  Production Regenerative Cycle
    Test Number                              1                   2
    Test Date
    %  Rated Output (58.6  mw base)           82.4                90.4
    NO  Controls                                     None
      A                •
    Water-Fuel Ratio
    Fuel                                             DF-2
   Stack Effluent
     Flow rate - Ib/sec
     Temperature - °F
    Water  vapor -  Vol. %                  4.52
    C02 -  Vol.  %  (dry)
    02 - Vol. % (dry)                    16.45                  16.15
  Visible Emissions - %  opacity
  Njtrogen Oxides  Emissions
    ppm (dry)                            232                     295
    pjpm @  15% 00    ...                   306                     366
    rf  "    •«  -2  w/EF                   199                     238
    Ib/hr                                 '—                     '"
 -Carbon Monoxi de  Emiss ions
    ppm  (dry)                            4.69                    4.01
    ppm 0 15% 0                          6.18                    4.96
              2
   Ib/hr                                 —    '
 Hydroca rbon Enn' s s i ons
   ppm (dry)
   ppm @
  Ib/hr
Notes:  (1) Average of 3 tests
        (2) Average of 2 tests
                                   C-73

    -------
                        REFERENCES FOR APPENDIX C
 1.   EPA test report  GT-8747-R, Rev.l, dated January 25, 1972, entitled,
     "Exhaust Emission' Test, Airesearch Aircraft Propulsion and
     Auxiliary Power  Gas Turbine Engines".
 2.   Letter dated February  13, 1976 from D. G. Medigovich, Airesearch
     Manufacturing Company  of Arizona, to Don R. Goodwin, EPA, and
     accompanying status report dated January 30, 1976.
 3.   Letter dated January 22, 1976, and coded ES:JMH:0401:012276, from
     J.  M.  Haasis, Airesearch, to E. A. Noble, EPA.
 4.   Letter dated August 26, 1974, from R. Kress, Solar, to Don R. Goodwin,
     EPA.
 5.   Report, "General  Motors Response to Preliminary (Draft) Proposed
     Standards for Control  of Air Pollution from Stationary Gas Turbines",
     dated  March 21,  1973,  and corrected March 28,  1973.
 6.   Letter dated October 17, 1972, from W. P. Slichter, BelT La~bora~tories
     to Don R. Goodwin, EPA, and accompanying test  report.
 7.   Sawyer's Gas Turbine Catalog, 1974 edition.
 8.   Letter dated April 23, 1976 from E. A. Noble,  EPA, to R.  Kress, Solar,
     documenting information provided verbally by A. Finklestein on
     March  26, 1976.
 9.   Letter dated November  10, 1975 from 0. M. Sievert, Solar, to Don R. Goodwin,
     EPA, and accompanying  submittal.
10.   NATO)  Engineering Data Sheet 799-052-003, dated December  20, 1971, on
     the Viking KG2-3 Exhaust Emission Analysis  (Diesel #2 fuel).
11.   Record of telephone conversation with Bill Wittner of North American
     Turbine Company  on September 12, 1972.
                                       C-74

    -------
12.   Record of telephone  conversation  with  S.  Lombardo of  Curtis-Wright
     on May 4, 1972.
13.   Record of telephone  conversation  with  S.  Lombardo of  Curtis-Wright
     on September 7,  1972.
14.   Letter dated April  18,  1972,  from R. W.  Wheeler, Alyeska Pipeline
     Service Company, to  T.  Kittleman, EPA, and accompanying  data.
15.   Letter dated May 11, 1972,  from R.  M.  Cummings, Alyeska  Pipeline
     Service Company, to  J.  A.  Eddinger, EPA,  and accompanying data.
16.   Test Report GTT-21  from Cooper-Bessemer,  entitled,  "Exhaust Gas
     Emission Test, RT-125 Gas  Turbine Package MO-280RP, Alyeska Pipeline
     Service Company."
17.   Test Report No.  11  from the San Diego  Gas and Electric Company entitled,
     "Report on the Sampling and Analysis of Emissions  from Kearny Mesa
     Turbine GT-2B."
18.   EPA Test Report 73-TRB-2,  "San Diego Gas and Electric Company Kearny
     Mesa Gas Turbine, San Diego, California", dated March, 1973.
19.   Letter dated October 16, 1972, from 0. J. Ortega,  Southern California
     Edison Company,  to Don R.  Goodwin, EPA.
20.   Letter dated November 10,  1972, from 0. J. Ortega,  Southern California
     Edison Company, to Don R.  Goodwin, EPA.
21.   ASME Publication 72-GT-53, "Nitric Oxide Abatement in Heavy Duty
     Gas Turbine Combustors by Means of Aerodynamics and Water Injection",
     by M. B. Hilt and R. H. Johnson,  General Electric Company.
22.   ASME Publication 72-JPG-GT-2, "Recent  Field Tests for Control of Exhaust
     Emissions from a 35-MW Gas Turbine", by M. J. Ambrose and E. S.  Obidinski,
     Westinghouse Electric Company.
23.   Appendix 3.a 1-2 of letter dated January 8, 1976,  from S. M. DeCorso,
     Westinghouse Electric Corporation, to  Don R. Goodwin, EPA.
                                    C-75

    -------
24.  ASME Publication 72-GT-22,  "Formation  and  Control of Oxides of Nitrogen
     Emissions from Gas Turbine  Combustor Systems",  P. P. Singh, W. E.  Young,
     and M.  J. Ambrose, Westinghouse Electric Company.
25.  Appendix 3.a 1-1, of letter dated January  8,  1976,  from  S. M. DeCorso,
     Westinghouse Electric Corporation, to  Don  R.  Goodwin,  EPA.
26.  Report from Engelhard Industries Division  of  Engelhard Minerals  and
     Chemical Corporation, entitled, "Catalytically-Supported Thermal  Combustion
     for Emission Control", no date.
27.  NASA Report CR-134736, "Experimental Clean Combustor Program, Phase  1,
     Final  Report", by the Pratt & Whitney  Division  of United Technologies
     Corporation and dated October, 1975.
28.  EPA test report 73-TRB-l, "San Diego Gas  & Electric Company South Bay  Gas
     Turbine, San Diego, California", dated March, 1973.
29.  ASME Publication 75-PWR-22, "Gas Turbine  Emissions  and Performance on
     Methanol Fuel," by R. D.  Klapatch, TP&M.
30.  MASA report NASACR 134737,  "Experimental  Clean  Combustor Program,
     Phase 1 Final Report", by the General  Electric  Company and undated.
31.  Letter dated January 9, 1976, from R.  H.  Gaylord, Turbodyne  Corporation,
     to Don R. Goodwin, EPA, and enclosures.
32.  Record of telephone .call on  February  5, 1976,  from E. A. Noble,
     EPA, to H. Gaylord, Turbodyne division of Worthington  Turbine International.
33.  Handout on emissions from an FPC unit  at  Bartow, Florida, provided by
     E. W.  Zeltmann, General Electric, at a  meeting  with EPA  personnel  on
     August 19, 1975.
34.  Letter dated October 31, 1975, from E. Zeltmann, General Electric Company,
     to K.  R. Durkee, EPA, and enclosures.
35.  ASME Publication 75-GT-68,  "Exhaust Emissions from  a 25-MW Gas  Turbine
     Firing Heavy and light Distillate Fuel  Oils  and Natural Gas",  D. E. Carl,
     E. S.  Obidinski and C. A. Jersey, Westinghouse  Electric  Corporation.
                                     C-76

    -------
                                   APPENDIX D
                      Emission Measurement and Monitoring
D.I  Emission Measurement Methods
     The new source performance standard for gas turbines was based upon the
results of tests conducted by gas turbine users and manufacturers and by EPA.
A careful review of the available gas turbine user and manufacturer test
reports showed that the sampling and instrumental analysis procedures used
were, generally speaking, based on the gas turbine test methods most frequently
recommended in the literature, i.e., the EPA Mobile Sources Jet Engine Test
      1                                                     2
Method  and the SAE Aerospace Recommended Practice (ARP1256) .   For this reason,
the test results were accepted as reliable and were used as part of the data
base for the standard.
     The EPA tests were done using a working draft of an instrumental method for
stationary gas turbines.  Information from the two literature methods cited
above were used in the development of this method.  The basic requirements of
the instrumental method are:  (1) performance specifications for the measurement
system;  (2) a sampling traverse to canvass emissions from the stack; (3) an 02
correction to adjust for variations in excess air.
     In all, EPA conducted three gas turbine NSPS tests; in each instance, the
instrumental method was compared against the manual EPA methods for NO  and 07
                                                                      A      £
(i.e., Methods 7 and 3, respectively).  The instrumental analyzers used in the
tests were calibrated with vendor-certified gas mixtures, which were reanalyzed
by the appropriate wet chemical methods.  Although zero and calibration drift
1  Federal Register. July 17, 1973, Vol. 38, #136, part 2, p. 19088-19103.  Control
   of Air Pollution from Aircraft and Aircraft Engines.  Emission Standards and
   Test Procedures for Aircraft.
2
   Society of Automotive Engineers.  Procedure for the Continuous Sampling and
   Measurement of Gaseous Emissions from Aircraft Turbine Engines.  (ARP1256)
   issued 10-1-71.
                                 D-l

    -------
                                         2
data for the analyzers were not reported, the calibration of the instruments
was checked frequently during the test program.   Sulfur dioxide emissions were
not actually measured during the test series; rather,  the sulfur content of the
fuel was determined by analysis, and the theoretical .SCL level  was calculated.
The calculation procedure is considered to be at least as accurate as actual
SCL measurement.
     The results of the EPA tests showed that the precision of the instrumental
data was better than Method 7.  Precision was estimated from the standard
deviation of the data sets.  In two of the EPA tests  the instrumental data agreed
to within 3 ppm of the Method 7 data.  However,  the standard deviation for
instrumental data was less than 2 ppm and for Method  7 approximately 11 ppm.
     After the test program, Method 20 - "Determination of Nitrogen Oxide, Sulfur
Dioxide, and Oxygen Emissions from Stationary Gas Turbines," was developed from
the working draft.  Quantitative values for the Method 20 performance specifications
were established during the test program using procedures described in the
continuous monitoring regulations of 40CFR60.  Recently, EPA validated the proce-
dures of Method 20, by conducting a simulated performance test on a controlled
gas turbine.
D.2  Continuous Monitoring
     Although  EPA has established performance specifications for continuous
monitoring  instruments in Appendix B of 40 CFR part 60, the specifications were
developed with the understanding that the instruments would be used to monitor
large industrial sources which operate continuously and have well developed velocity
and temperature profiles.  Gas turbines do not fit into this source category.
Furthermore, EPA has  not, to date, conducted any tests to determine performance

                                 D-2

    -------
                                        3
specifications for monitoring this type of source.   Therefore,  EPA cannot
recommend continuous monitoring instruments for gas turbines.
     Additional factors which could affect the feasibility of  continuous monitoring
of gas turbine emissions are:
     a.  Gas stratification found in some turbine exhaust stream makes location
of the sampling point a critical  factor.
     b.  There is a lack of personnel at remote turbines to routinely check and
maintain the sampling equipment.
     c.  High costs are associated with emission monitoring at turbine facilities
because a single site may have multiple turbines.  For example, the costs for
opacity monitoring have been estimated $20,000 capital and $8,500 annual operating.
Since an opacity monitor is an in-stack type monitor, each turbine would require
a separate system; it would not be possible to time-share its  use at adjacent
turbines, as it would be with an  extractive type unit.  The costs of gaseous
pollutant monitors for NO  and Op are estimated at $30,000 capital (per monitor)
and $16,000 annual operating.  Facilities which have more than one turbine could
reduce this total cost by selecting an extractive type system  and by time-sharing
the system between several turbines.
D.3  Performance Test Method
     Method 20 - "Determination of Nitrogen Oxide, Sulfur Dioxide, and Oxygen
Emissions from Stationary Gas Turbines," is recommended as the performance test
                                     1       2
method.  The use of the Mobile Source  or SAE  test procedures was considered but
was rejected because these methods specify the use of particular types of instru-
ments.  However, both the SAE and Mobile Source test methods are acceptable
alternative methods, if the selected instrument models meet the performance
                                 D-3

    -------
                                         4
specifications of Method 20.
  .   The selection of Method  20 for performance  testing  was  based  on  the results
of EPA gas turbine field tests (Section  D.1).   Method 20  includes the  following:
(1) measurement system design criteria;  (2) measurement  system performance
specifications (including analyzer span  drift, zero  drift,  linearity  check,
response time and interference checks) and  performance test procedures; and
(3) procedures for emission sampling.
     EPA Method 6 is recommended as the  performance  test method for SOp.  The
sample point location for Method 6 is  prescribed in  Method  20.  The sample volume
of Method 6 is not fixed and  may be increased if the expected SO^  concentration
is very small.  In lieu of measuring S02 the  standard permits the  measurement of
the sulfur content of the fuel by the  applicable ASTM method.  The sulfur content
is then used to calculate the SO* concentration.
     The cost of conducting a Method 20  emission test is estimated at $4000 to
$6000 per turbine.  The testing costs per turbine may be lower If several turbines
at a single site are to DQ tested.
                                 D-4

    -------
                  APPENDIX E - AIR QUALITY ANALYSES

Dispersion analyses have been conducted for nine Individual  units  and
for several cluster arrangements as are discussed 1n  Section 6.1.
Concentrations are estimated at distances of 0.1, 0.2,  0.5,  2.0 and
20 kilometers from the downwind edge of the units.  Because  of the
large volume of data, the results are reported for only one  set of
emission levels.  For the analyses of Individual turbines  and the
cluster arrangements having all turbines operating in the  same mode,
concentration estimates for other emission rates change proportionately.
                               E-l

    -------



cu
en
«
!_
0'
^

CM
O
CO

r—
rtJ
3
C
c
CJ
co
CO
en

OQ
p?J
•
u '—•
CO
C3^ Ez
^
U> CO CO 00 ^fr"
to ^. ... •
**-*• O> CO r— O
fc"^
00
*
o
                                                                0)
                                                                Cn
                                                                re
                                                          o
                                                          cu
                                                          co

                                                          en


                                                          CM
                                                           •

                                                          n
                                                                r-    g.
                                                                (O
                                                                C
                                                                      un
                                                                      c\j
                                                                             CD
                                                                             a.
                                                                            CO
                                                                              •

                                                                            CM
                                                    CO    x
en en
(0
t- 00
0) CO
>
u co"*
CM E
ocr'^^r^-Looco
3 ^- ^- LO CO O
j- co — ' en co >—
3
O M
ac co
1
«* o
CM






en
CU
8
^j
i-
^3
O
1
CO







u
CO
CO
en
<*
u -—
CO
ex E

S. 3
f%

CM
CO







 E
•r*
 X
       

00
CO
 II

cr


CO


00
 •
o
                 co
e


3
                                     VO
                        CO
                               o
                               U)
                        CM    «S-
                        CM    i—
CM    r—
  •      f

      CM
                      s,
                      
                     o
                     o
                                                     o
                            o
                            0)
                            CO

                            O)
                                                           II   CO

                                                          cr
                                                                             en
CM    r—
  •     •

O    O
                                                          CM
                                                          co
             CU
             u

             10
            4->
             (/)
            •r—
            O

            •o
             c

            I
            i—    CM
                                           o
                                            •
                                           CM
                                                 o
                                                 CM
                                  E-2
                                                                       r—    CM
                                                                                   LO
                                                           o

                                                           CM
                                                                                               o
                                                                                               CM

    -------
                                   o
                                OJ O
                                17) (/I
                                (0 ^»
                                fc. CD
                                (U
                                > CO
                                «£ CO

                                 CM '
                                O II
                                CO
                                 O CO


                                *f 00
                                CM  •
                                 O     CO CO iO CO CM
                                  £      •   •  •  •  •
                                 -^,    IO O CO VO r-
                                  O>    •— VD CO CO
                                  a    CM r- r-
                     vo
                     cr>
                      x
                      (U
                      I/)
                      (O
                            u
                            Ol
                         ^
                         (O en

                         O> CO
                         > oo   x-
                         
              0)  S-
              O  Ol
              E -P
              O  
              O  3
                      (U
                      >  -u
                      
                                 C  U

                                "55
                                 C -t-)

                                 842
                                00
                                       •— CM
                                                o o
                                                CM O
                                                   CM
                                                           E-3

    -------
                             o
                             a)
                                                                                     u
  oi   en
  en
  re  co
  1-  CO
  0)    •    —
  >  I—   CO
 <:         E
      n    >•
  CM        C
 O  Or     1.
 i—  CO
 re
 3  *<
 c  oo
 c
 <  o
                                              oa    o
                                               •     •
                                              r—    CM
             oo
              •
             o
                                                                                  0  •


                                                                                  < II
                                                                                         co
                                                                     en
                                                                     a.
                                                                                                 O    r-
                                                                                                                  VD
                                                                                                                    •

                                                                                                                  O
                                                                                  (O Q-
                                                                                  3
                                                                                  C 10
                                                                                  C CM
                                                                                 < r-
                                                                                                                                     o

                                                                                                                                      c
                                                                                                                                      to
                                                                                                                                      V)
                                                                                                                                      V)
            Ul
CO
 a;
jQ
 re
                  V

                  re

                 4J
                                                                                                                                      £
                                                                                                                                     •»J

                                                                                                                                      0)
                                                                                                                                      u

                                                                                                                                      O
     o
     QJ
     .
 en  en
 (O
 i.   co
 
                                       in
                                       CO
o    co
  •      •
CO    i—
CM    CM
                                                         in
                                                           •
                                                         in
                                                               oo
                 8    j.  «
                 r_    CM  O
                                                                                     u
                                                                                     Q)
                                                                                  0) CVJ
                                                                                  O>i—
                                                                                  IO  •

                                                                                  cu r~
                                                                                  > II
                                                                                  3  Q.
                                                                                  O
                                                                                 a: o
                                                                                  i  in
                                                                                 oo CM
                                                                                         CO
                 >


            *   J
                 E

                 _E

                 X
                 re
     u
     O)
                       at   en
                       en
                       re   co
                       j-   co
                               CO
  CM
O   cr
co

 s-   co

 o   »«
31   00

CO   O
                                       oo    •—    en    en
                                 O)
                                 a
                                       00    r-
                                       in    r-*
                            CM    VD
                            CO    r—
                                        CO
                                                                                     o
                                                                                     o>
                                                                                  0* CM
                                                                                  Oil—
                                                                                  re   •


                                                                                   u
                                   o
                                                                                  i  in
                                                                                f— CM
CM    r-
  •     *

O    O
                           
    -------








"












«3-
VO
0"!
r—

M «•*••*
to i—
ro 1
x vo
a,
s_
** 13
to en
* £
_1 'ro CU
LU O CU
zj to
1 1 • n *•- -~*
LJ_ • " ^«i**
c
LU O S-
«* Z -r- CU
1 LU -(-> -4->
1 IX. CO fO 
«: o i- 3
.~*> _J C£ 4-> r—
i— O LU C O
/-» WJ i^ CU
f8 U -4->
|_ • " C T-
|cS i

h- r- CM
< (U CM
CU TJ
_J C
•o
c in
o
i.
CD

E
3
E
•r—
X
*o
3E!



















O
CU
CU to - — «
CD -v. CO
(O GO CO E
c 1/7 *^^
cu a* • co
> co ii a
«a; . o* » — •
CM 0
o cu
oo to
3 CO CO
o oo co E
1 *< r — CO
«* 00 II 3.
CM • cr —
U
cu
to •— ••
cu -^. co
co 0) E
i. • CO
CU »< II 3-
> co cr —
«=c •
o
CM CU
O to
00 >>*
CO *~-«
$_ co co
31 ?\ cf\ cr
\J i \ t c
o • ^
a: a* i — co
i oo it a
co • cr •—*








cu
en
ro
S-
Q) U
> CU
. co
3 a. . ^
ac co n co
ooJSo J-

S-
a>
«4-J
(/)
^3
!•«
O
cu
CO 4->
ro • i—
i. 0 C
cu cu r>
> to
»
3: co co
• vo E
r-,_ CT >— •
T3 CU
C 0
•r™ CI ^"^
2 (O E
c 4-> 22.
Is ~




o oo co in
O i— 00 00 O
VO OO GO r™1*
CM r-




O O O 00 «*
CM CM O 00 r—
O"> 00 VO ^"
VO ^- CM




O O O r— CTl
00 CO CM r— O
VT> CU O CO
Co m CM






o o co r^ "^*
in r— rx CM CM
in en co oo
vo co un
CM i—



















a> vo co r-
* • • •
oo o o





















VO CT> "* r—
• • • *
r- O O O




r— CM in o o
CM O
CM




CM in vo co r-
cn vo co *i- c
in <*• CM





m vo oo **• <*
t^ CO CM i— C
lf> CM VO r-
r— r™




o o r-» CM VN
vo •— **• r-. c
COCO"*
CM i—






O O O CM IT
P^. CT> CT> CO C
CM »*• i— r—
VO CO i—




















CM,-- |
* ' • 1
00 0 |


s-
cu

to
3

O

1 *
,^»
c


in





CO CMr-
• • *
O O O




r— CM in O O
CM O
CM
                                          vt
                                          £
                                          •M


                                          §
                                         t_J
E-5

    -------
              o
              CO
 co en
 01
 id OO
 i- in
 co   •
 > co

   ii
  CM
O cr
oo

r— OO
 to
 3*«
 C 00
 c   •
<0
                  CO
                         O   i—    •—    r-
                                   U
                                   CO
                                   00

                                co en
                                CO

                                s_  •
                                CO CM

                                < II

                                _xt=r
                                                               id  Ot

                                                               c in
                                                               c CM
                                                                      CO
                                                                         o>
                                                                          •
                                                                         o
                                              r>>
                                                •
                                              o
CD

O
                                                                                                                     O












«
X
£

id
i— CO
r- CO
id id
«i f^ &••
UJ CO
£ c <
n



o
00
^^
CO

00
LO
OO *^~N
oo












ui .^ evil E
in
i „
* §
•g *"
*~
Z 4J 0
LU UJ 00
§-P S-
C =>
uj 5 o
.« c -}-
Of 3 CM
cr o>
— ' CM
00
00
o
•* r— P»> CM
• * • •
 ii
                                                              o
                                                              o

                                                               s-  S.
                                                               3  a.
                                                               o
                                                              3= O

                                                              00 CM
                                                                      CO
                                                              CO  •—   •—
                                                              E
                                                             •-^  o   o
    E
   •r—
    X
CO CO
en
>d co
i- in
CO  •
> co    ^-~
<:     co
   II     E
 CM      ^
oo-
00
           $- CO
          3= 00
           I
          CO O
                    o»
                    3.
                         en
               10
               CO
co
CH
                                    in
                                    OO
00

VD
CM
                                                oo
                                                        o
                                                        CO
                                                        (/I

                                                        CO

                                                     co in
                                                     coo
                                                     Id   •
                                                     S. CO
                                                     CO
                                                     >  II
                                                     3  a.
                                                     o
                                                    2C O
                                                     i in
                                                    •— CM
                                                                      CO
          •o  co
           C  U
          •r-  C
           3E  tO
           C •!->
           X  (/)
               •—   CVJ
                                o

                                CM
                                                                               in
                                                o
                                                CM
                                                                o
                                                                  •

                                                                CM
                                                                                          O
                                                                                          CM
                                                         E-6

    -------






~-

































VO



^"CL)
^
JO
£Z

• A
o:

^£
i—
UJ
o





















































•5f
VO
en
r-*
*
CO
to
X
O)
1—

ft
fO
r—
r—
0
(— J
• M
C
o
.f~.
•M
ro
S-
4J
C.
0)
(J
c
0
o

r*_
0>
o>
—I

•o
c
3
0
i-
CD

E
E
•r-
X

^
3
CO
lL

O)
O)
I/)
-*»-*•

t-
at
•4-*
to
13
^^
o

-u
•r-
C

CM
CM

-o

ttJ

LO





























a>
CO
(O
>
^
o
oo
^
0
T"
1
CM








ai
n\
\ji
to
S-
a>
>
^^

0
0
00
i-
^
0

co















01
en
S-
0)
>
^c

0
o

S-
13
o
^
oo




O)
CO
S-
O)
>
^^

o
o

S-
3
o
1
^
T3
C
•^
c





OO
CO








oo
*R
00







oo

%A
co
J •








oo

^•t;
00
•


















cz
o
ex

o
LO
CM













E
a.
Q.
o
LO
CM
ai
(J
c
•u

o
0>
-^c.
CO
CO
II
cy

0
CD
CO

en
00
LO
oo
II
cr
o
ai
{/)


•3-
CO
•
r-™
ii
Cr


o
0)
CO
-^
co
00
LO
•
CO
II
o-













o
a>
CO

en

LO
CD
•
CO
II
o-








o
ai
CO
-^^
CO

LO
o
co
ii
Cy







*~*
E
CO
"*~^





fvi
\ t
E

CO
^





f^\
VT/
E

o>
^.
x»^







rrt
" t
E

co
3.
x.^



















^—v
co
E

en
^












^-^H
co
E
cn
^E_




P=
.1

































































S-
O)
+J
CO
3
O

•4->
C
ra

CM
CM














CD
00
00
r~






0
,_
o
LO






o
CM
O
0
r—








O
CM
^^
vo
CM























VC

C















00
•—




•—




0 O 0 VO COr— 1
. . . .
ooo 1















O CM CO 1
i— O O 1




CM LO O O
CM O
CM


CM OCM i— LO
r^ co ^ CM o
CO «ef CM r—







CM O tO CO CO
V7> ^" 'sf' CM r—
CT> i— VO CO
•—






O O OC!> ^"
i — LO CM ^1" O
CM VO VO r—
CM r-









O O O LO CVJ
cn o LO !*>. r—
oo •«»• vo co
LO «* l—
























9~~ r~
• •
^ 00
ai
4J
CO
3
J3

4->
'^
^^

LO




LO COr-
O CDO




i— CM LO o a
CM d
CsJ
                              O
                              i
                               0)
                              •M
E-7

    -------
VO
          0)
                u
                cu
             oo to
                cn
             o co
           CM
          O
          oo
                Cr
i—     O
    r—
                                  CO   If)
                                    •     •

                                  •—   O
                     CO
                      £
                      •^
                      cn
CM    CO


--'    CD
 cu
 O)
 10

 ; DF-2 FUE
icentration
08
o
in •—
coo.
cu
•o
c
1
CO
Maximum










on
o
cu

a* cn
cw o co
cn •
2-
cu
>
«*
CM
0
00
&_
3
O OO
re
1 S*
** OO
CM •







OO
CU
05 S<

 r- - O ' **•
*^^ • • • • •
cn r>. co r- r** tj-
3. CO IO CM



CO
\ •— i— CO •— CO
cn • • . . .
3. r- «3 VD CM r—
~— • "ST r—







^-N
CO
E O co vo r>~
"**. • • • . .
cn in co ix> co *j-
^3 °^ ° cn CM CM





oo
E o co cn o »*
"•^ • • • . .
5 5 S g ^ *
•>— ' r—





CU
cn
2
cu o
3: ""
< w»
0 CT
E • oo
J- Q.^- E
3 Q.CM \
O O)
n: o ii E CM r—
i in •£, . . i
CO CM CD- O O 1








CU
cn
IO
cu u
> cu
^
t/» *>••*»
0 CO
j_ I.-; E
*» Iwla'VJ ^^^
3 Q.CM cn
fg" ^ «? *i -.
r— CM CD1 O O O



                                                                                                                         •4-)
                                                                                                                          C
                                                                                                                          CU
                                                                                                                          u
                                                                                                                          c
                                                                                                                          o
                                                                                                                         o
        •o cu
        c o
                          f-    CM
                                   O
                                     •

                                   CM
                                                                                 i—    CM
                                                   O
                                                   CM
                                   O

                                   CM
                                                        E-8
                                          o
                                          CM

    -------
u
(U
cu cn
CD
03 00
C. t f\
^ *— V) •
2> ^^ If—
>
O CD
i oo co
CM** cn
O CO
• II
CO O
cn cu
cu cn
CO CO
co •""' ^ r—
<00 II
•> cu • cy
CO S- CM
<0 3 O
i— cn oo
I— •!- O
(0 U_ i- CU
a 3 LO
cu o ^
_i •« cu z cn
O LU C LO 1
i— « ZD O * — CO 00 CO
co i— 4J i- ^ cn
O CM ro CU O CO
1 LU 1 i_ +J • ||
—1 LL.4J CO r— O"
s— tU LO Q C 3
CO r—
CU _l •« O O
•— <, o_ c
-Q C2 •— 0 •*-»
tO LU OO -f-
1— Z O C
LU Lflr- => 0*
(S C/J CU 0>
Z > LO ra
0) r- i-
_I CU
-o >
•0 E <
C tl O
3 O CU
O " O co
i. 00 E -^
eg i.o.ot
3 CX
E •* O ^1-
§ a: o CM
.^ 00 CM Or
X
s:
cu
cn
(O
i-
cu
0
o cu
O co
t-Q.cn
3 Q.
o «*
3C O CM
Ilrt II
•* * II
i— CM cy


^~/ -o cu
C 0
C +j
3: co
ss.



^*~+,
CO
^
CD
^







co
^
cn




^— ^
oo
^
cn
J3_
* —







il*" ^
CO
£
o>

"^^^















^•^*
co
E
•*»^
Cf>
O-









co
E LO

cn-i-
c er
J-i
VO
r~


"E"
—





O O O O CO
oo «* o «cr 10
CT> CO 00 CO
CM i—







o o o o o
LO LO LO CO i—
CM "Sj- LO i — CM
CTi «* CM r-




O O O O LO
r^. ou o CM ^™
co i— t^ o i—
tx ** CM r—









O O O O LO
«* CO CT> CT> 00
LO r~» cn co co
^ OO CO CO
CM r—



















CO LO «el- •— O
r— O O O O












^" r— t*^ CO O
CM r- O O O




r— CM LO O O
CM O
CM








































































(/)
4->
•r»
C

00





 O O O CO •—

 P-» LO CM LO CO
 CM O «d" r-
 CM r—
                        O O O r— LO

                        LO CTl CO 00 r—
                        CO LO •—
O O O r— LO

LO CM O ^t O
LO LO «*• LO r—
r^* co r—
O O O O CO

<• LO co r^ LO
LO CTl «sf CM
                                                           LO
                                                           w
                                                           cu
                                                           CO
                                                           C
                                                           o
                                                          •fj
                                                           C
                                                           cu
                                                           u
                                                           C
                                                           o
                                                          o
 ooo or-*

 co co i— cx> LO
 co LO «d- ^~
 r**- co r~
                        o o o o cn

                        CO LO CO O CM
                        co cy> t^ LO
o o o o •—

2; LO o **• CT)

CM r—
                        O CM O O VO

                        r^ oo "?f r— cn
                        LO «?f Sf 00
                        ^r LO CM
CM LO CM i—

r-^ O O O
                        r^ co i— o o
                        o o o o o
CM    0 0
•— CM LO O O

         CM O
            CM
                        CM LO CM O O

                        i— O O O O
                        r- CM LO O O

                                 CM O
                                    CM
   E-9

    -------
            QJ
            to
           i—   CO
      CO OO  •     E
      Cn   CO    -^
<:  *"
       o
  CM   O>
O     
00    ^.
       Cn
r—    CO

                                                                                                 4=
                       CM
                             co    o
                                               10
                                                                                                 c
                                                                                                 o
                                                                                                 •r-
                                                                                                 -p
                                                                                                 H3
                                                                                                 •»->
                                                                                                 g
                                                                                                 o
                                                                                                 o
ro
X
V



^
0)
•o
1 /"^
0
I-H
OS
O
LU
_J
LU
j$
QC
LU
LU
5^
3
U. 01
^
CM S-
1 OJ
U_ M
O QJ
C£
• M
0. cn
r— C
§'r-
c
in c
oU.,.
oo











«
c/t

o
r*~
0)
>
0)
•o
c
i
CD
E
*3
•^"
X
fi



o> to
a>
ia fc*
J- O
QJ .
> r—
<
CM
O
00
s-
3
O 00
1 **
 E
CO" ^ * «> * «"»
« a CM ^- u> co
Cj >-* CO r—











O
0)
I/)
^v.
cn *~*
r- CO
                                                         en
                                                         1-
                                                         Ol
                                                               O
                                                        o
                                                        O    r-
                                                           £  •
                                                        i- J5.CO
                                                                  co
                                                        coincy
                                                                    ff
                                              «
                     cn

                                                       g
                  s
                                  -

                                                                                     2    5   S
I
  0)
  o
  C


 S
 •r—
                                                                         r-CV,U,
                                                                                           0
                                                                                            •
                                                                                           CM
                                                                                                O
                                              E-10

    -------






















^~
LO
cn
•—
A
to
to
X

*~
•*
to ^*" **
fO r—
^ 10
it)
O CU
s_
_l •« 3
o LU tr cn
•-H =3 O T-
CC U-T- UL,
H~ 1 *
O CM «3 0>
LU 1 S- (U
LU O C 	
CU
_l ••> 0 S-
«t O_ C 0)
cBr- 0 -<->
LU OCO tO
as o 3
LU LOl— r—
CS CO CU O
CU 4->
1 »f—
c
T> ID
c
3 LO
O r—
t-
CD
E
3
E
X
. ^_>r
s_
CU i—-
tO 0)
a» T3
Of 0
cn
c s~
C i —
C 3
•r- CJ)
a. a>
co oe
U 0
cu cu
to to
cn cn
cu -— -
cn co co co
fa co cn co • E
t • | i "s^^
cu a* co fc« i — cn
> CO II CO || 3.
«t . o- • o- ^
cTc* o z
CO CO CU S U
10 CO
i- **>. to
3 CT) ^v
o cn •• — »
ai i— co co co
i • • E
CM Oi— O CO OJ
• II -II 3-
r-crr-cr —
cc 2;
co S

o o
CU CO
\ ^
cn cn
CU '~~-
cn co oo co
10 co cn co • E
cu frs co fc« i — cn
> co ii co U p.
 to
CO WJ >v^
> **». cn
< E cn E — *
Q.I— D.LO CO
O Q. • Q.t». E
o co >^.
O^l- oo o cn
S- O II • ll E

o
i c2 i
oo co 52

cu
cn o
re o cu
S- CO to
cu to ^.
> \ cn
< E cn E ^ S
0.1— o.ir> co or S
o a. • o.r^ E co
cj co • **^ to
o ^- oo o cn to 4-»
S_ O II • II £:•*->•'—
3LOOT^.Cr 	 -r-C

3; .« .« =
1 CC 2 CM
r— CO S *f i—



•a cu
c o
•r- C .—»
3 . «=t LO
10 CM r~- co
CM r-







o o o o o
cn co oo CM oo
CO CM LO i — i —
00 *J- CM r—









CO O O O 10
^ r~» o LO cn
oo LO LO cn
LO CO CM








o o o o o
to cn ^~ . co
CM r-







o o o o o
CO 1^ O i— LO
cn o LO i— •—
r^. <* CM i—









O O O O 00
r— LO cn co f***
co co ro cn
LO CO CM








o o o o •—
O LO tO r~~ LO
P*. r— Cn O CM
F** r— r*-» CO
^— f—















o to »-«3- I
. . • . 1
tO CM •— O 1










ry y
co ce.

CO CO CO CM CM LO I
1 * 1 > • • • • 1
•r- 'r— t"^ ^ i— O I
c c
~» — ^

00 00





i— CM LO O O
CM O
CM







O O O t CO
co cn co co CM
«* CM. h- CO
CM i—







O O O O 00
i— CM ^t- r— r-.
r- co *i- •—
00 •* CM i—









0 0 0 0 CM
r— LO cn oo ^i*
CO CO CO f*^
tO CO CM








o o o o o
o vo LO cn sr
f"% r— cn LO t—
r-. i— r^ CM
















LO cn oo oo 1
• • « • 1
LO CM r— O 1









a: s


to en

c c r-^r CM r— o



LO O





r— CM LO O O
CM O
CM

















r—

o
c
*c
•p
V

a
r*"
v

c
•^
4-
It
4-
c
Q
t.
c
c
t_




















































E-n

    -------
vo
            00
       u
       <1>
       co

       Cn
            co
         >
         <
       II
      cr
 CM
o

1/1     o
i—     OJ
 it)     co
 3tO\
 c     en
<0   •
    • cn
   r— CO
       II
                    co
                      CT
                      3.
           CO
             E

             cn
                                   r—    O
                                                             cn
                                                             to
                                                                      x
                                                                     o
   u
       VO
                                     ...
                                   10    p.    ,_
   in
E  •    E
Q.CO    ~^.
Q.CM    cn
         3-
                                                                               co
                                                                                     co
                                                                                       •
                                                                                     CO
                                  cn
                                    .
                                  O
O

 C
                                                                                                                              w»

                                                                                                                              8
               0
               0)
               CO




1
LjJ
(U

X]
ro
»—






























CJ
S
CJ
UJ
«J
UJ

*E
on
UJ
UJ
CO




























UJ
£
™
u.
a
• *»
CO
CM
8
to
CQ
3^.

























A
r—
Q
• A
C
O
•r—
-M
i-
-«.,_>
C
(j
c
o
0
r-~
(U
>
0)
TJ
3
O
i.
cs
E
3
E

X
(O
2:













(O •
>
g
a>co
> •


CM
O
CO

s^
3
£
1

^O
•
t^


w ro O CM co «3- at
00 £ • . . . .
* --«» CM CO r- ^- i—
r~ 3. r—
II
Cr


OJ
CO f *
O) E O O 00 00 iO
CO **^^ • • • • •
CM p. r>. ^ CO r—

II
{~Y






(J
(U
CO <— »
•*«. co o o ^ in to
cn E • • • • •
CO *^» GO IO *^ ^ r^
cn r— co ^ i— i—
cn 3. ^f »—
co « —
II
c^



o
(U
to ^-~~ O O O it- t~^
— CO .....
cn E co co co oo co
co ^ cn in ««j- ro co
• Cn CO "Sf r—
CT> 3 r—
CO 	
II
ex
verage
*c
8
j-
3
O
1
CO











01
cn
re


«c

0
o

i.
3
o
1







0
to
cn
Q.-*
a. CM
O ||
in
CM cr














u
at
to

cn

E .
a.**
a. CM

in







                                                                                                                              a>
                                                                                                                              o
                                                                                 CO
                                                                                              CM    •—
                                                                                CO
                                                                                        •—    a
        •c a»
         c o
        •r- C
         2 *
         C •!->
         5 CO
             E      r-
                          CVJ
                                in
                                      o
                                        •
                                      CM
               r-    CM
                            in
                                                      o
                                                      CM
                                  o
                                    •
                                  CM
                                                                                                        O
                                                                                                        CM
                                                           E-12

    -------




































v_>
h— 1
CM oe
i— h-
o
I UJ
UJ UJ
01 _J
S S
it) Ul
1- 2
UJ
C3























































^j.
VO
Cfi

CO
rO
01
1— ^
„ 1
CO ^°
<« QJ
•— i-
r— 3
(O U^
Q T-
U.
_J •«
UJ C QJ
Z3 O OJ
U. •!- CO
^ i ss.^
CM 03
1 S- J-
Q E +J
 rs
0)
—1 CM

•0 TJ
C C
3 (0
O
S- -el-
ID «
00
E

£
•r—
X
*O
^

















^v
en
co •—*
• CO
QJ CO r— E
fO ft^i Cft
i- CO II 3.
co . cr —
^
CM
O
CO O
QJ
i- CO
3 "v.
o en
-I- ,— ^
i coco co
_4_ ^
^^ C
^1 ^ft O^ ^^^
C^3 O^ ^^
• II 3.

O
QJ
CO
en
oo • — •
• CO
en r— *>x
ftj ^^ C^
i- CO || 3.
QJ . Cr —
>
<^

CM
O
CO
0
i- QJ
3 CO
O -v.
rn en
i
CO CO *•"•*
co • co
*« CO ~^.
o en
• II 3
r- cr ^-*





01
en
to o
S- CD
0) CO
^ **^
*C E en *•""'*
o CO
O Q.«3- E
O CM \
o en
S- LO II E
3 CM O" - —
o
a:
i
CO
QJ
O1
CO
i- 0
QJ QJ
> CO

a. en co co
O a. E *->
o CM en E
*- LO || £ HD
3 CM CT > —
O CO
.J—
1
-O QJ
c o
^10 ^
§•*-* **
CO 	
ss



o o ^i- vo r^.
O O O r- CO
CM rx oo co
r»- CM









O O O O LO
O O 00 LO CM
O O VO O i—
«* (TV CM i—
CM




o o o en r>«.
r*^ oo ^D r*^ vo
r>» LO vo oo
r- 00 CM
CM












O O O O LO
LO O 00 CO CM
LO VO VO CM CM
CM 00 CO CM
fx CM















VO *fr «* r— O
.... .
CO i— O O O










,

en ^t i~x CM o
LO CM O O O




r— CM LO O O
CM O
CM



O O CM 00 CO
00  i—
O i— LO < i»>. CM 5
CO CM ,C
4-
(T
i
4-
C
a
t
C
o o en o c
• • i • ^^ t.
r- r- en CM •
f~» CO VO CM r—
r>» CM












O O O <$• VO
o en co co, LO
^» vo co f*^
LO CO CM
CM












,


CO ** r— i
...
r— O O 1










CO

E r— S. CM I
CM O O 1
CM



i— CM LO O O
CM O
CM
E-13

    -------
0
CO
to
-^
CO CO
COCO CM
03
t-VlCQ
0) O CO
> • II
<«-o-
CM
O
CO
r— O
ft) CO
3 in



















*J-
<^>
CT>
•»
(/)
IQ
X
CO
c
5
c

it
c







CO
^
CO
in
o •
vo
•« CM
0 II
• CD-


0
CO
CO
^^
CO
1— COCO CM



A
to
i. *•

e oo
 O) O 06
(O > • ii




CJ
«. *— '
CO ry
i — (_
CJ
1 UJ
LU UJ
0} ,
s s
£ z
ill
UJ
CD
























'~
5
o
1
«J '
Ul ••>.
^ c
U- CO O
CM O 4J
1 >> 
*^
•«*• d>
CM in
to •
vo
*«CM
CO II
« O"





CO
coco
(tj
1- S*
CO O
> .
«c •—
CM
o
co
1-
o
nil
1
CO
co
a^s
co
•
T3 CO
c u
•i- C
2 *
C 4->
2 co
o £



u
o>
tt\
VI
o>
CM
•
00
00
II
cr



o
<1)
10
co
LO
VO
CM
II
CD-






CO^
E
a ". ^ ". ". °°
•^ •— O O CD CD*

co
E
\
» <*> 0 0 0 CM
** * • • *
-— 0 0000






1^™x
CO
E^^ . -fc _^
^^ *-| VO IV. fv. o
s s ui ifi * ^
v^ CM






CO
E
"^ «> CO PS. ^- ,_
-3 S « * r-*  CO CO
r""** |-^





^^
CO
-f, °. "1 •-;'*•*
? °° f^: °° o o
^ c^ "> ~ - -


"E1
•^ •— CM LO o O
*"* * * • • •
CM O
CM
                  co    r-
 CO
 (O

 cu

<     o

O     10
°  r-"^
    E  CO
 s-  cxvo
 3  O.  .
 O    CO
a: o «a-
                       CO
                              10    r—
                                •      •
                              o    o
                              o
                                «
                              o
    o
      •
    CD
o
  •
o
                              r-    CM
                                          in
                                                o

                                                CM
                                                      o
                                                      CM
E-14

    -------
o




























o
c2
,_ 1 —
o
1 ' ' *
_J
LU LU
0) — '
•— S
4:1 LU
iJ? Z
•"" LU
o















































vo
cn
^
M
cu T
1 — VO

C/) t-
(O 3
r— Oi

gil
CU
— 1 .« CU
LU C (/>
2 -IB *
CU -M S-
CM •— « CU
1 O ^ -A*
L|_ >fc 4r.* I/)
00 C 3
••» CU o O
CO r— g;
5 f=" ^ *"*
O -i- c
r*** co r— """»
on cu
jf-^ ^^ lf^
CU T—
	 1

T5 C

^m
^3 ^
S-» OO
ts

E •*
E
X
0}


















CU OT
(13 LO CO
CU LO ^
> S« CM C7)
•»
O co cn
' " v3
u
O)
cu cn
ftj LO CO
cu vo ^^^
> ^^ CM cn
< CO II 3.
. o- —
CM
O
oo
o
i- CU
3 C/>
0 --^
•XL en
co to CM co^
*« CO ^^
O CO Oi
• II 3-
•— cr —





cu
O)
(O
i-
cu
> 0
«=c cu
(/)
o -^
o cn
S- S.LO CO
3 O. • E
o co ^
oo CM cr *~"


cu
i.
cu
> o
<: cu
O -~v.
t_5 f*^
S- Q.LO CO
3 CX • E
o co ^
• in u E
r- CM CT 	 •
"O 0)
C (J
•i- c:

C -M E
Tf t/) \X

o o

o cnr^^j-«>t-
CO VO i— r— r—
r-* Cn ^" r— r—
CM







o o QI cn
§00 cn oo P^
CM CO CO CO
f^». CO r~



o o o vo cn
LO in r>. r— CM
fs^ ^Q LO "^ CM
P^> CO !••






O O O O LO
vo r~- LO cn LO
co r— CM co r*.
in CM in r—
CM i—


















O *t CM i—
r-^ O OO








-M 	 	
•1—
c

VO CM r— O O O
""




r— CM in o o
CM O
CM

t~~i ^. cn cn r^
vo co ^- «* m
r»» vo CM
r~







O O O VO r—
in cn co vo cn
CO CM 00 f— i—
in CM



O O CO VO VO
in oo ^ cn i —
co m cn ,r—
VO CM






o o o o in
00 O ^t CM CO
r™ LO r— CO CO
r- CO CO
CM


















00 CO r— O O
O O O O O








4J
>r_
c vo r>»oo f— .
•— o o o -1
CO




r— CM in O O
CM O
CM

o CM cn co co
LO lO CO CM CM
CO-"* r—
•~







O O CM LO LO
LO LO <*
•*r-



O O VO VO CO
oo cn CM co m
oo vo in







o o o co co
i~x n LO co cn
CM VO r*. CM f—
vo in i—
^w


















*°* ^ *"• 1 1
0 00 1 1








10
•r-
C CVJLO CM
^ •••II
i— O 0 1 |





i— CM LO O O
CM O
CM
                      E-15

    -------
                               u
       «/)   »—.
    tst «v.  co
  41    O.    E
  O *U CO   <.
   I CO CO    O

      '«   e
                          L.
                          CU
                                          CM
                        VO
                         •
                       0
                                                     CM
                                                       •
                                                     O
       CSi   ^-
        •     •
       o   o
                          CM
                         5?
•—    cu
     --^
                           co -x.  co

                         cuw#   ^    r   °I
                         CDCO 00     C;    00   CO
                         «  •   •     a    in   i-
                         L.   00    «-*
                         CU    II
                         >   Cr
                                                     r*.   co    r-
                                                      •     •     •
                                                     iet   r—    i—
                                                   C
                                                   0)
                                                                                 u
                                                                                 CU
if>
LU

CU

J3
n   ui

     ui
     o
                          CM
           v  y  £.
          r~-  >  ft
       CM  U  »-  *J
        i  >, a»  ?
       u-o  w  h
R

 3i
 o
                               ai
                               
                                  CO
 I O   •    C7>   CT)
«a-   • o>    a   i—
CM r— CM    •—
                 O    •—
                   •      •
                       CO
                                                    CO
      o
        •
      «o
to
 •
CO
 3 OOO


fg"
00 U» Cr
                                                                                        .
                                                                                       O>
                                                                                                 If)
                                                                                                   •
                                                                                                 o
                                                                               CM
                                                                                •
                                                                               o
                 2
                C9

                 (=
                 3
      U
      01
      l/t   *-
 41 V? ^v  CO
 01   cn    P

 2^^
                        0)
                        >
                         CM
                        o
                        j_
                        3
                        O
     00
       •
     oo
      U
      CU
      V)
                                    3.   •—
                                              CM
                                              VO
                                                    CM
                                 VO
                                                   £
                                                   I
                                                                                    co
   00 ^   -~~*   O
 I     O)  CO
CO *« ^-    £    »—
   o  •   *s»   eh
    • o    en   to
   •— CM    3.
      II    —
      or
CO

s
                                                         o
                                                         CM
                                       o
                                       CM
           3 O.CO     CXi

           f§"     *
           r-in cr
                                                                                            CO
                                                                                             •

                                                                                            CM
                      CO
                        •
                      o
CO    •—   r-
 •     •     •
POO
                           a>
                          5
                r—    CM
                                                         o
                                                           •
                                                         CM
                                                                        CM
                                      O
                                      CM
                                                                                                            O
                                                                                                              •

                                                                                                            CM
                                                                                                                  o
                                                                                                                  CM
                                                                E-16

    -------























«^-
VO
O^
r— '

•S
Jo
X
cu
^«—
#>

10'—
r— 1
i— vo
(O
o cu
S-
_l "3
OLi 1 g~ QT)
UJ w. wi
i— i :O O -r-
VO Q£ |j_ > S- Ol
V^ _J U. C? +-> CO
1JJ 111 Q C V^
cu cu
0> _l ."r- 0 i.
rr rf on ex c co
•£> 2 i— E O 4J
ro LU O T- O CO
1— Z O 00 3
uj rv. r- r-
CU OO Ol O
y >
O1 4J
_J 1-
c
•o ID
c
3 VO
O r—
i-

E
i
•r—
X
z:

















s —















































































rv
oo
V— *s~~*
^f
cu o;
> > 	 ^
s_
Ol r-
CO CU
Ol "O
cm o
o>
C S-
•i- to
C r-
C~ -5
•r- CD
Q- Ol
oo c£
o
U Ol
CU CO
-v, en
CD LO CO
to IS) CO ^O * E
Ol a* 00 »* CM CD
> CO II CO || 3.
«c • o- • cr ^"
^\J • *\ * A
o as o as; o
OO 00 CU S Q)
CO CO
£— ^^ ^^^
3 CD CD
O **~^
z oo ^ oo CM co
)• • ^^
• • tz
^- *5 CT> ** OO ~v>
CM O CM O CO CD
• II • II 3.
i— Cfr— cy *—*
• ts • A

oo S
U
U O)
Ol co
CO \
"»s. CD
O) CD 	 ,
CD LO CO
i. • oo vo \
01 &« CO fc« CM CD
> CO II CO || 3.
< • o- • cx -^
CM-" •«
o cc: o z: o
oo oo cu clS o)
CO CO

3 CD CD
o ^~-
a: oo ^i- oo CM co
O CM OOO CD
• II • II 3.


ck z;
oo cr




cu
CD O O
ro Ot CU
S- CO c/)
Ol -^ — .
> CT) CD
< £ £ '—•
0.00 ci.cn co
O Q. • Q.CO E
0 •* *v.
o co in o a>
rj LO O*CM cy — •
o
i ct: z:
oo oo S


0)
CD O O
(O QJ Ol
S- CO CO
O> ^v. *>^
< E E ^
O.CO £XC7> CO
o a. • Q.CO E
(_>•«*• . «»y.
CD CO LO C^ ' CD
S- O II • II E
3 in cr CM cr — •
o
^T" . «v • *i
i a: z:
r— OO ^****




-o cu
c o
• r- C . — »
2 ro E
Cjj ^7
•T-* . Jfc
5 co • —
O -f-
Q Q




o o co o r^.
o en fv. vo en
CM O •* r-
CM r-







O O O CO ^t
i— LO 00 CO CM
co vo in in co
IV.. CO r—










O O O LO *J-
^± ^- en co en
co oo vo LO r-
r-v co >—







o o o o vo
t*o r— «3" 00 ^J"
^- CO VO tv, VO
*s- CM in F—
CM r-
















VOLO intM o
CO r- O O O










ctrz:
oo S
en vo en CM i
•(-> 4J in CM CD O '

E c
^^ !^D

^- CM
f.M






r— CM LO O O
CM O
CM





o o m co o
•— 00 r— O OO
CO r— in CM
CM i—







CM o o vo en
o LO CM rv. vo
rv. en r-v vo CM
rv co t—










o o o co oo
^j- co oo «*• in
LO O tv. VO r—
^^ *5jf" (•" •







o o o o rv.
«* CM CO ** CM
i— «* en r— in
in co in CM
CM r—
















en co o> co
*J- CM O O










%%
co CM rvio •—
+j +-> co si- r— a o

E c


00 00







r— CM in O O
CM O
CM





O O Qi—
VD O CO CO
«» co in CM
CM r—







O O O VO
en CM CM co
r— co en en
CO ^d" i*"^










O O ^~ CO
LO VO O CM
CO CM CD 00
fv. v^ r^"







o o o o
in i— tcj -sj-
r— CM CO rv
vo ^ vo 04
CM r—
















co 'j' m rv.
VO CO r— O










ex z;
to a£
CO CO 00 i — CO CM
1-1- O VO CM r—
C C i—


VO
r— O







i— CM in o
CM





r"
c

1
1
J

I
1
1
r"

c
£
{
•r
4
(
!
4
c
c
t
c
c
c.


































































E-17

    -------
                           
                                   (/)
                                                                             2
                                                                             0)
                                                                                    u u
                                                                                     CO   •—*
                          o           co
                          to    CM CM    E
                                  • oo   ^
                                  i oo    en

                                ii  ii    —
                                                                                      i O)  CO
                       1o   co
                                                in

                                                CO
                                                          co
                                                           •
                                                          evj
                                                           r-    O
                                                                                    Q.  CO ««*•
                                                                                  i  Q.  «a-  .
                                                                                  i      u  u
                                                                                  : co    to Q.
                                                                                          e
                                                                                         •—

                                                                                          3.
                                                                                                           00
                                                                                                                 C\J
                                                                                                  CO
                                                                                                             10

                                                                                                             o
                                                                                                                             in
                                                                                                                               •
                                                                                                                             o
            IO
            en
            X
            0)
UJ
O
UJ
    UJ
^s
It)
UI


UJ
   CM C
_^<-> £
loo    *j
       §1
   8

£q
CO
 0)
 2>    O O
^f.    QJ A)
      io co
 CM   >-^^
f^    J^^ J^   ^^

CO           CO
      CM CM    E
 S.     •00-^


 O    00  •    -
Z CO H  II    •*-

 i      cm a.
                          CVJ r—
                                              CM

                                              s
                                                en
                                                  •
                                                o»
                                                CM
                                                         co
                                                               CM
                                                                                 CU
                                                                                 D>
                                                                                 (O
                                                                             8
                                                                                        o  u
                                                                                        0)  0)
                                                                                        (/)  V)


                                                                                        O> O)
                                                          in   o
                                                          CM   r—
       *x> 'i?
         &• CO
       CO «*
a n.  «*  •

a: o    co a.
 i  m
oo CM
                                                                                                 'I     i     I      I
            a*
            X
            |Q
                       01
                       cn
                        cn
                      _
                     oo
                     o
                            CM   _.  _.
                          O           CO
                          CO    CM CM    E
                                 • CO   \
                           &•    OO 00    C7>
                           3    oo  •    a
                           O CO  II   II
                          co
                                           in   in
                                           in
                                           oo
                                                    CM    •—
                                                    oo    CM
                                                            CM   CTt
                                                                 CO
                                                                                 0)
                                                                                 o>

                                                                                 2
                                                                       8
                                                                        *.£
                                                                                         o u
                                                                                         Q) a;
                                                                                         co co

                                                                                         en en

                                                                                        *o co
                                                                                          • CO
                                                                                        CO V
                                                                                           co
                                                                             O     II  II    ^
                                                                             *iC ^j    C^7bX>   C7)

                                                                             •-^ CM  crcr3  —
                  u co
         CO CO
            to
         c a.
                              
                           5  to
                                              i—    CM
                                                         in
                                                            o
                                                             •

                                                            CM
                                                                                                      •--   CM
                                                                 o
                                                                 CM
                                                                                                                       o
                                                                                                                         *

                                                                                                                       CM
                                                                                                                             O
                                                                                                                             CM
                                                                E-18

    -------
























C3
t—*
-o &
. t—
0
-! LU
_J
LU LU

OJ -^
s 2
* LU
*~ LU
C0









































«3-
vo
O
r—

A
to
ro
X
CU
r—

A
t/1^*
*rt
^T
i— US
ro
CU Q CU
i— J-
_J O •« 3
LU >> C Cn
EO O i-
•I- LL.
t3 -M
CM CU ro CD
1 C J- CU
U_-i— -M CO
OXI C^^
§03
U i-
QQO C 
C CO 4J
=>_!••-
1^
3^0
o •—
i.
IS

E

E
T—
X
ro
SI


^-^.
Q.
CO
^-~^l^— ^
co
z: -^
*-^ 0
ro
-i^ -t->
C3 CO
fO
4-> CO
CO to
ro
C Q-

'ro >,
z: oo





CU
en
ro
CU
< o o
CU CO
CM to  U O
 CU CU

00 CM &QT -^



CU
en
ro
t-
a> u u
> ai cu

E -co ' — -
1- O.CO •<* CO
3 CX^j- . E
OHM -^.
3: o COD. en
i m atco E
c— CM rcro* — '








-o cu
C 0
•r- C *-^
5 ro E
C •(-> H
2 to •—'
ss


o oo •*»• i-~
• • • • •
F>» CM O Ch O
CM CM O 00 LO
•a- co CM













O O O O ^~
CM CT> CO ,— • CM
•* •* «* O CO
00 O 00 CO
















CO CM i— i—
• • • • 1
o o o o 1

















r*. r>- yj CM .
• • • • 1
oooo 1












i— CM If) O O
CM O
CM
                              CO
                              10
                              to
                              C
                              o
                              (O

                             •«->
                              C
                              CU
                              o
                              C
                              o
                             o
E-19

    -------
             o
             HI
             to

             en
         co
                 co
      o
<     CO

  CM  -x.
O     cn
      co   *-
r—    co  co
            Cn
       3


       E= CO  ||
      
o
(U
                              00
      cn
      3.
                  VO
                  CM
                       O    GO


                       CM    i—
      m
       •

      O
                                                           93
                                                           Cn

                                                           £
                                                           0)
                                                            X
                                                           o
                                                                 u
      cn
      3
                        en
                       O
                         •

                       CM
                              U3


                              O
in
  •

o
CM
  •

O
   i01
r— CL  •
 
oo ^*-
co »« ^d-
cno vo


O \
co en
i_
3
O
2=
^.
CM

CO
CO
OO •
cn
CO ||
• cr

co
cn
*~*»
co
•v^
en
3


O
•
O
in
in

0
in
to


o
vo
00
CM

CO
m
CO


cn r^
CO 00
in i—

CM IO
to in



to
CO*

10
CM



                                                           QJ
                                                           en
                                                           ro

                                                           a*
                                                           o
                                                           cj
                                                                                                        2

                                                                                                        i
                                                                                                        u
                                                                                                        o
                                                                                                        o
                                                                     CO
                                                                 en
                                                                E

                                                                cn
                                                                            CO    r-
                                                                              •      •

                                                                            o    o
                                                                                  I     I
                                                              CLCNJ
                                                          0:0

                                                          00 CM i
 0)
 O


O
•r-

 X
            u
            a>
            en
                 *—
         is> *t  co
 coa< «i-
 cno 10
 >o   •  u
 t.!— cr
      o

      o •»-
      a a
          i—   CM
                      10
                            o
                              •

                            CM
                                  o
                                  CM
                                                      E-20
                                                                     r-    CM
                                                                                 in
                                                                                       o

                                                                                       CM
                                                                                             o
                                                                                             CM

    -------



_-.



























o
CM 0
1 UJ
LJ —1

CU
£ §>
(!E ui
z

CD























•-~-'



























VO
cn
^
ro
X
CU

CO CU
3
r- Cn
(O 'I-
ou.
iH, *" •» co
y*3 c  i ^>
QT; c co
CU 3
^T ?* 
0) VO
™"^
T3 T3
c c
3 1C
o
i- «
CD OO
M
3


X
ar




















o
0)
CO
ju ^ r-»
Cn Cn *^J
g E
cu co • cn
> cn 3.
co n
CM • or
o
CO
fc.
3
o o
X CU
1 CO

c\j cn co
co ^t- ^«.
cn

o vo ^— ^
• u
i— O1

o
CU
CO
CU ^ -—
Cn D^ ^O
fO c£
s- ro ^.
W co • cn
> cn 3.
CO IT
CM • O-
o
CO

>- o
•3 CU
O CO
ac ^. -— «
i cn co
oo E
CO "3- "^
cn
O VO **-**
• II
r- O-




CU
cn
(O
s*
CU
> o
< CU
CO
o ^». *-»
o cn co
3 Q. • Cn
0 CM E
HI O ^ ~— '
i in u
00 CM O"

CU
cn
 o
< CU
CO
O ^ •*"-
o cn oo
u£* ^
3 Q. • Cn
0 £+ ^.
i in it
•— CM or
-a cu
c o
•t— C
S - in vo CM
CM i—





O O O O CO
^^ 1^ ^-P ^^ O^
vo co o co i—
Si- «* VO CM
CM r-





O O OO •—
o cn o r— vo
O r^ CM CO vo
CM r-. o r>.
00 st CM













CM in out t
. . . . i
Sf CM r-O 1













CO
-M
•r- rx CM cn r^« r~
Z3 VO **• i— OO
vo




r- CM in O O
CM O
CM



o o o «*in
in o o coin
CO O r— sf
«* CO r-









O o oo in
1^. O VO CM 00
VO O CO r- """
l~~ H"~





oo oo cn
CM "d- co co cn
vo co m r—
f*^ Cn CO r"





0 O OO CO
Si- CM CM VO OO
r». r— cn cn co
CO i— r- CO
in co i—













cn vo vo CM
CM i-^ VO O














CO
4J VO N, r— *!•
C «* CM i— O

CO



i— CM in o o
CM O
CM



o o f"^ CM r»»
r^ r— cnr- CM
o co in CM
•* r-









O O 0 r>. CM
r^. CM cn o cn
ino cn i*-.
OO 03 i—
r~~





o oo sr o
in h<- CM cn in
co in cn in
r— in r—





o o o o i-»
CM r-» o co vo
oo in «* cn r-
f** CO *O r—
CO i—













cn cn co r— .1.
• • • • i
r— 0 0 O |














CO ,
+j oi VO VOCM j
C CM i-^ OO 1

"*



r- CM in O O
OJ O
CM
o

 c


I



 CU
 £


 CU
 u

 o
u

    -------
      Ol
 • II
< •— Or
CM
O
to
r- O
10 CU
3 in
C ^
c: en
^-^
CO
E

3
N.^







•— X



in co o co ^j-
• • • • •
r— CO *t CO O
^- p.








   oo
      3
   00  II
           co
            en
            3.
                                                                                                            O

                                                                                                             c
                              CNJ
                                    o    i—
                              l—    r—    O


cu

en to







o
»— i
C£
CM J^
UJ JIJ
^3 fly
J? IAJ
>~ z
3









C
O
•r—
+J
n>
. — * s-
0) -P
_J 0) "O C
UJ i— Q 0>
=3 o ac u
1  (Ui—
f"j »f— (/) (y
4J CU >
• •> fo DC cu
CO S. _l
i— Oj en
O C C T>
O CU -r- C
f^ O) C 3
^ Cj C O
a. co
CO
•*— • *• £~
g
•^-
X
IT3
3T
^.
O)
>
^c

o
0

i.
3
o
T
CM











O)

o
•
I—

0





t^
CO
•











cnoo
in








i-
cu
>

S-?
o
•
r—
u
m
•
CM
II
^5"





CU
en
10
cy






o
a>
(/)
^x»
en
in
,_!
CM
H


co~*
E o o o ^- cy>
en co i— CM r— *t
3 ^- oo in co
*— IT) i—






^-v
CO
E O CM VO «tt ^3-
**x^ • • • • •
CT> CO «^- tO CFi ^~
a vo ir> r-











^M^
co o o co CT> cn
-^ 
•~s»
en

«^-
vo
u







^™^t
CO
E O O ^ 00 r-
^^ • • • • •
cn <^- r— oo in f^.
3 O CM CO i—



cu

(O
s_
(LI
< <->

S ^
E W *~~*
$- O-CO CO
^T" f^t QO Cn ^5j" J5J* *^ r~~ r~
I CO It E * • * • •
co m cy •— * co r— o o c)














O)
cn
i.
cu

«C 0
0)
O V)
0 "^
E Cn *— ^
J- Q.OO CO
3 Q. • E
O ^ "^.
moco cncM i— r-»co co
i o ji E • • • • •
•— incr — -m CM o o o
c
•r"
4-
K
i
c
8
g
c«





































                  CO    r—
T3 O)
 C. O
•<- E
                  •—    CM
                              in
                                    o

                                    CM*
                                          o
                                          CM
CVJ
            o
              •
            CM
                                                                                                    O
                                                                                                    CM
                                          E-22

    -------








*-~-





























o
HH
Q£
^~
C\J O
r LU
i
V_- UJ

CU _J
r"~ *^
«O *"ir

LUr- C CU
r?° o s-

O -Jj cn
CM (tf T-
1 CU S- U_

Q-r- C CU
4-> CU CU
•«MTJ O V)
cog c«^
r— JO o
OE o s-
Ocu CU

CO CO CO (/)
2-OC > 3
CU i —
_l O

"O 4->
C "i—
3 C
O ^3
13 tO
c "
3
E
X
rO





































































































o:
oo
	 *
cu s
^ '"*'*
i- •• — "
cu
t/) Ol
CO TJJ
0,0
cn
c: s-
•i- to
C 3
•r— CO
CL CU
oo on
cn coco OOOIDCM
*3" CO -•». O r— CO CM CT>
oo . O CM ID CM
«* CO CD CM r—
co •* oo • P.
^^ • CT^ ^-.x
re> co u co u
s_ . cr • cr
cu
< (^ 5
co cr
CM O <_)
o cu cu
co 10 to
--x. --^ *~* OOO^3-CM
3 E ^ O tO tO ID
OOOincO^f ~~^ «±<5j-lDr^
n: • C7i oo CM oo r-.
CO O CM O to — *
• II • II
r— Cfr- CT

• *> • **
oi y*
co S








cu
cn u
ID CU U
s- to cu
co -^ to •— »
> cn ^^ co
 cn ^
< E cn -^
Q.CO co a: ac m ^a- in r-«.
c^ f^ • P r**^ E oo *^f • • • •
O «* Q-CTv ^» O tO CM O
o oo a. • cn to to r-
S- O II II E -t-> 4J
0 EC
3C •• => =>
i a: z;
i— OO S «* CM
|«*



TD CU
C O
•r- c: • — • r— CM in o o
C -l-> -2 CM O
5 to 	 CM
O -r-
a a
o o o CM m
o> to to cr> r^

r-^  m CQ & ir>
cn co co to CM
co co in CM
CM i—






O O O O ^~
in CM in in co
to i-^ r— cn r—
CO r— ID r—
r— r— "










O O O O ID
i^ to to cn «3-
i— O r— <* «t
CM cn r^ to
IO CO r-





















«* ^}- CD O • —
CM r»» CM r- O
r—












Q£ aE *a- O ID COr—
r^ i — «* e— o
(/> r— i —
•!-> *->
C C
r> ro

00 CO






r- CM ID O O
CM O
CM


o o o>— o
O> tO Sf i — *f
CO O tO Is*
vo in r—






0 0 O O •*
cy^ r^» r^* r^ co
cn co •* co ^~
CM to in CM
CM r—






o c o o cn
CO OO ID •— tO
t-v «a- si- in
to O «3- r-
r*" r—










O O O O CM
CM CO ^f ID CO
Ch P^ Cp O CM
tn  to i —
to r-^ r^. CM o
to to CM r—

c c
=> 5

to o






r— CM ID O O
CM O
CM













•"
c

,
f
J

t
r"

i
.J
4
r
S
•4

I

<
*•





























































E-23

    -------
O)
cnco
to
i- M
at o
>_
•

£r
0
a>
V)
S
•
r>.
CM
n
cr
CO
E
Cn C^ r— r— r- 00
•— • , C O O O
CO
E
en i— 1 I I CM
^ •
-^ 0 10

s>
2
I
u
x a>
O 00
Zi« -
r— CLCTl CO
« DL • E
3 VO -^
c **• vo en
c vo n a

      CO
                   co
 o> o en    en
 >   •  n     a.
«c.— cr    «—-
        CO
       oo
         •

       00
       01
                                CM
                                      CO
                                       •

                                      vo
fx

  •


**•
01
  •

vo



CO
CM
1
LU
01
fM>
•Q
£
F"*


0
fr-4
oc
b
LU
_l
LU
.
* en
CM CO
*
co r>.
»« CM
CO ||
• cr







CO
E
•^ vo *
en •
a cr> ^
^ CM









J- Ol ^J- r-
• • • »
•» r— i— CM




                                                        
                                                        i. O.P*
                                                        3 Q.  .
                                                        O    O
                                                        x o to
                                                                           CO
                                                                                   CM
                                                                                    •

                                                                                   O
               u
               o>
 E
•i—
 X
                 •  II     3.
CC r- O    >^
O
CO


 t-

 3

 O
       O   O   O


            SCO   CM

       L«   "*"   *°
       LD   r—   i—
CO
  •

VO
CO
CO



 en
                          O

                          in
                                CO
                                      10
o>
  •

o
CM
  •

O
 Ol
 O)
 (O


 I
<    u
      o>

8    -^
    E en
 i.  pLr^
 3  BI .
 o    o
a: o vo
 i  in u
•—csicy
                                                                           CD
            (U
            (J
                          r—    CM
                                      IT)
                                           O
                                             •

                                           CM
                                                                                 CM
                                                 O
                                                 CM
                                                                                             O
                                                                                              *

                                                                                             CM
                                                                                                   O
                                                                                                   CM
                                                      E-24

    -------
0)
































(_)
t— 1
cc
o
LU
_J
LU
i
s
LU
LU
CD














































*3-
LO
CT>

^
t/)

— ' O QJ
1 1 ^ w
"-•r- to
_.-(-> — •
2s«-
fe-P  =>
0)
—1 IO

T3 '~
C TO
3 c
O (O
t.
o oo
A
E<*
^E
X
(O

















o
0)
 &« CM cn
«t CO II 3.
• o- -^
^N 1
v Vl
O
CO O
O)
S- M
3 "^
o cn
1 CO CO CO
^J- • E
CM »S CM -^
O cri cn
• II 3.
r- O- 	



O
0)
(/)
~^,
O) cn
cn *—N
tJ 00 CO
t- co • E
 *« CM cn
«C CO II 3-
• o- — •
CM
o
co
(J
t- Ol
3 to
o \
X cn
CO CO CO CO
*« CM' -^
O cn cn
• II 3.
r- O- ^


O)
cn
73
i.
o>
> o
<: QJ
t/1
o -^
o cn
s- cxr«« co
30.- E
0 0 <
X o LO cn
' LO II £
oo CM cy • — •


0)
en
 o
cj; QJ
if.
Vl
o ^
t-) cn to
E -^ -M
S- cxr^ co "r-
30.- E C
00^3
"T" Q VQ Q}
1 If) || £ IO
•— CM O- — - i—
E «
•S °
s§ J
1= -



o r^ LO LO o
CO LO •— LT> r—
O «3- CM r—








O o r»» r^. r^
CM LO i— 00 LO
^ LO r»* i— co
CO i—








o o o cn o>
LO CO LO CO CM
O f^ i— CM CM
LO CM i—







O O O 00 CO
i — O 1** Ol VO
CM •— oo r*» r^
o cn co
CM













r-. co r— | I
... i i
o o o 1 |













r- OLO •— •-
CM •— O OO


r— CM LO O O
CM O
CM



LO Cv. if) «3- LO
SCM CM r— LO
COr-








O O LO 00 LO
CM Ch r— *t 00
00 O «d- i—
CM i—








O O "* CO LO
CO «1- CD Or-
O CT> LO r— r-
LO r~







O O O CO **
LO r*. CM ^*- oo
^ ^t1 CO CO PO
LO LO CM













If) CM»— I 1
* * * 1 i
o o o 1 1









(/)
4J
r™

=> rs. co co | |
•••II
oo r~°° ' '


r— CM LO O O
CM O
CM



cn in oo O r»»
LO •— LO r— CM
LD CM







f
o r>. oo LO CM
o r- CM co cn f
CM C*. CM ' J
CM 4
i
t
«
f«

<
1
1
+
o o r*. r^ f^ i
CM 00 00 ^T LO C
Ch CM CO (
CO •— I
(
c





o o o cr> CM
LO i^. cn LO cn
O CM CM r- r~
co •«• i—













*^l 1 1
o o 1 1 1










trt
L»
T**
E CO lf)r- 1 I
= ••'II
^ r- 00 I |


r— CM IO O O
CM O
CM
                                                E-25

    -------
    vo
                   o
                   Ol
                   t/>   •-
                OO "•». CO
                   O.   E
              £r*    ~
                             \o
                               •
                             CM
                                             co
                   H
                   O
              CM
             O
             «—    O
              fO    01    «—-
              3    «/»   CO
              C 10-^    £
              c    a-    <.
             <*?co    a>
                   u
                  o
                                       oo
                                        •
                                       o
                      CM   •—   i—
                       •    •     •
                      o   o   o
     in
     ra
     X
     ai
     I/)
     IQ
                             v
                             v\
                      CO
                                             r-    CO
                                               •     •
                                             ^t    CO
 5  .0"    a   ^
 >r- CO   «—
                                 01

                                 §
                                 Ol
                                                                             u
                                                                             0)
             0~
              i.
              3
                   U
                      CO
                             CO
I
u

u
a
a
                                             CM
                                                                              "*•  co
                                                                           Cr-  '  E
                                                                           B-CM   -^
                                                              X O  H
                                                               1 O
                                                              oouncr
                                                                                        1O
                                                                                             in   CM
                                                                              ,-   o
••   0
                   U
                   Ol
 &»<<
 «0
 L.  • i
 Ol r— I
 >
<
                      CO
                                        CO
                                             Ov    VO
                        Oi
                        3.
                             O
                             o
     CM    ^-   u>    r*.
     rv.    r»   ^-    r-
in   r—
                   it
                  or
                                                                        Ol
                                                                        O)
                                                              <

                                                              8
                                                                    •  CO
                            u
                            01
               : «/>
             co
               co  .
                  T
                                 I
                                      CM
                                       •
                                      in
                                                              3 ca.»-    o>
                                                              son    *~"
                                                                              CM
             •O 01
             c u

             IS
             X (/)
                                  E    —
                                            CM
                                                       o
                                                        •
                                                       CM
                                                                                       O
                                                                                        •
                                                                                       O
                                                     E-26

    -------






__






























o
VO 1—
CM O
1 UJ

Ul
v^^

j^ ^f
» O r—
O** O *« CM ^ O CO O> CO CO
rnocoooi 01 incM
i • u • it n.
«*r- OP- CT --^
CM
o: s
co a:
0
cf U CO
<2> co in
Ol --. CO Ol CM CO P—
x en cr> »« c\4 01 in CM r—
CO-—- (O »* II CO II 3-
>_T IjCrjO- -0- —
re co ^^ S
r— i- CM
r- 3 O 0 0
re ot co co co
a •!- 10 wi
U. I- "*-w 	 	
_J .. 3 01 01 OOOOVO
ra o co 3: co oo co co co r-»omco*j-
u_ .;. u, i • -E ^i- co CM «— vo
jjO- COS i. 1. • II • II p^
u. ^> a> •— err- o" —
O C -4-J
CO «/i IT .IT

O c r- CO CC
r- O O
O o
O 4->
r^ r-M-
Ig « = a. u
fl) O> CO U
_j vo  •d-r-do
§^- J_0 u • u E
3tno-cM o- —
E O
« 1C • ** •**
x i o: 2:
re coco 52
z:
CO V->
O) CO 0
10 Wl CO
s- \ «/>
CO O) -^
> Ol
< E -sf E *-* aejE.
OL • Q.r— CO CO Qi
O Q-r- D-in E
. — . O CM • \ 
OH O r- i— O Ol -M +J Ol Ol O CO
	 3 in O^ CM O1 — ' C C VO CM r— O
•^ O 13 r3
> S 1 0£ S CM
1.^ r— CO 52 •* r-
0)
in co
S|
O) "O CO
C S- C O 	 .
•r- nj -r- C E
Cr— S•»-> m vo co vo p—
•p- E O> ^ r- O O
p 3

CO 00





r— CM in o o
CM O
CM

C r~" *^" 1^- CO
p— 00 CO Cy> tl"
Ol Ol "!f P"~
p**











0 00 «> •—
r>» r». LO m vo
CO CM «* «3 P-
iO CO r—









o o o n &
m in 01 p- oo
CO CO ^ LD
C£> CO r-?- ^







O O O O d!
VO VO VO «* Ol
r*- r— CJl p CM
P— r— SJ" tM
CM P-














oo in m r> r—
vo tr> i— o O











a:
1/1 5
in
•M 01 co CO r— CM cy
C M- CM VO CO r- O
—^ £: r~~
VO
p- O





r— CM in o o
CM O


E-27

    -------
                     APPENDIX F.  ENFORCEMENT ASPECTS

     Factors which affect gas turbine emissions are changes in power
output, fuel consumption, fuel analysis, ambient conditions (temperature,
pressure and humidity), and water or steam injection rate.  The effects
of these variables have been discussed in Chapter 3.
     Test method,    "Determination of Nitrogen Oxide and Oxygen from
Stationary Gas Turbines", specifies test methods and procedures for both
field and laboratory measurements.
     The compliance test for stationary gas turbines requires the use of
instruments to measure 0? and NO .  Instruments were specified because they
                        C~       J\
can accurately measure low concentrations of pollutants and are best suited
for production line testing.  For repetitive emission tests, such as during
engine performance testing, the stack can be designed so that one sample
point will be representative.
     Compliance testing of gas turbines using the methods specified in
the regulation should be straightforward.  The instruments may be based
on any of several analytical techniques which meet the performance criteria
specified in the test method.  Possible problems which may be encountered
are instrument malfunctions and stack configurations which are not amenable
to reproducible sampling.
     Testing of turbines may present a special problem which requires an
interpretation of the method by the enforcement agency.  EPA's reference
test method specifies that emissions from the turbine shall be sampled down-
stream of the power turbine.  Furthermore, for combined cycle plants which are
supplementally fired, sampling shall be performed between the power turbine
and the boiler (this excludes emissions from the supplemental firing).  In
                                   F-l

    -------
turbines which are ducted to a common  stack,  emissions  shall  be measured
in the individual breachings before the common  stack.   Less  satisfactory
alternatives such as testing at the ,top of the  stack,  upstream of  the
power turbine, or at a single point, should be  avoided.
     Modifications which may be necessary to  permit emission testing
may include installation of an extension to the stack  or temporary removal
of noise silencers.  Any modification  should  be reviewed by  the enforcement
agency to insure that it does not result in an  unrepresentative sample.
     Although emissions from a given gas turbine are reportedly almost constant,
they increase markedly with a decrease in overall  efficiency. A decrease
in efficiency can be caused by degradation of the combustor, erosion of
the compressor or turbine blades, or deposits on the compressor or turbine
blades.  For large turbines used to generate  peaking power and burning
distillate fuel, the rate of efficiency decrease has been estimated at about
one percent per year.   Changes in efficiency can be monitored by  standard
process instrumentation, such as pressure ratios, exhaust temperatures,
and fuel flow.  Combustor life can be  estimated so that maintenance activities
and emission measurement tests can be  scheduled accordingly.  Deposits of
material on the compressor and turbine surfaces are periodically  removed by
methods such as steam cleaning or by injecting crushed walnut shells or rice
hulls into the air intake.
     Compliance tests to verify emissions must be performed after  each major
repair and when combustor sections are modified or replaced.  If  the process
instrumentation indicates that a significant  degradation in  turbine efficiency
has occurred and engine maintenance or overhaul are not scheduled  for the
near future, a compliance test should  be performed to determine compliance,
and if necessary, provide a basis for  specifying mandatory maintenance to
reduce emissions.
                                   F-2

    -------
     Formation of NO  from organic or fuel-bound nitrogen is not retarded
                    J\


by wet control methods, but is retarded by some dry control  techniques



as discussed 1n Chapter 4.  The performance test must be performed while



the turbine is using a fuel which contains the maximum organic nitrogen



which will be used at the installation.



     The recommended emission levels permit the actual NO  emission rate to
                                                         A


be adjusted based on the efficiency of the gas turbine.  Those with greater



efficiencies (greater work per unit of fuel consumed) typically operate at



higher temperatures and pressures.  This is conducive to NO  formation and
                                                           /\


higher concentrations of NO  are emitted.
                           A


     Correction factors to adjust NO  emissions to a reference humidity and
                                    X


pressure have been recommended.  If a manufacturer has developed correction



factors for humidity and pressure specifically for his model of turbine,



he may use those correction factors after submittal of substantiating data



to EPA and approval by EPA.  Correction factors for temperature are optional



and, if used, must be developed by the manufacturer for the specific model.



The International Standards Organization  (ISO) standard day conditions of one



atmosphere, 59°F and 60 percent relative humidity are chosen as the reference



conditions.  Since the correction factors recommended were developed by



manufacturers for turbines which use conventional combustors operating at,



or near, stoichiometric conditions, they cannot be applied to turbines



which use emerging technology combustors such as the low emission lean-burn



types (described in Chapter 4) which operate at off-stoichiometric conditions.



Correction factors for turbines which use non-conventional combustors must



be developed by the manufacturer and approved by EPA for each specific turbine



model.
                                     F-3

    -------
     The supplementary fired portion of a combined cycle gas turbine,, when
fired with a fossil fuel at a heat input of greater than 250 million Btu
per hour, must meet the standards of performance for fossil fuel-fired steam
generators.  The gai, turbine portion of all combined cycle units shall meet
the standards of performance for gas turbines.
     Since the maximum emissions of NO  from gas turbines occur at full
                                      J\
load, compliance tests must be performed at full load.  Since NO  emissions
                                                                ^
will vary with changes in  ambient humidity and pressure (as discussed in
Chapter 3), gas turbine owners using water or steam injection "to"meet"the
 recommended  emission  level  referenced  to  ISO conditions will have  to  vary
 the water  flowrates commensurately.  An alternative to this  variation of
 water  flowrate  is  to  use  a  constant water flowrate which will  provide a
 reduction  in  uncontrolled NO   emissions to meet the recommended  level under
                            J\
 the worst  anticipated conditions  of ambient humidity  and pressure.
     In  arctic  and'subarctic  areas, the "ice fog" which sometimes  forms  may
 result in  reduced  visibility  and  traffic  problems  (see Chapter 4 for  discussion),
 The regulation  permits suspension of the  use of wet control  methods on those
 occasions  when  the resulting  ice  fog is deemed a visibility  hazard by an
 authorized representative of  the  state or local government agency  responsible
 for traffic  safety.
                                       F-4

    -------
                          REFERENCES  FOR APPENDIX  F

1.  Letter from Dlbelius,  N.  R.,  Vice Chairman  of  the  Combustion and
    Fuels  Committee,  Gas Turbine  Division of  the ASME,  to Steigerwald, B. J.,
    OAQPS, EPA.  p.  5.  May 1,  1973.
                                     F-5

    -------
                        APPENDIX - 6
      METHOD 20—DETERMINATION OF NITROGEN OXIDES,  SULFUR

  DIOXIDE, AND OXYGEN EMISSIONS FROM STATIONARY GAS TURBINES
1.   Principle and Applicability



     1.1  Principle.   A gas sample is continuously extracted  from



the exhaust stream of a stationary gas turbine; a portion  of  the



sample stream is conveyed to instrumental  analyzers for determination



of nitrogen oxides (NO ) and oxygen (09) content.  During  each NO
                      X               <-                          A


and 0? determination, a separate measurement of sulfur dioxide (S0?)



emissions is made, using Method 6, or its  equivalent.   The Op



determination is used to adjust the NO  and S09 to a reference
                                      A       £


condition.



     1.2  Applicability.  This method is applicable for the determina-



tion of nitrogen oxide, sulfur dioxide, and oxygen emissions  from

                         •

stationary gas turbines.  For the NO  and  0? determinations,  this
                                    A      C-


method includes:  (1) measurement system design criteria,  (?.} analyzer



performance specifications and performance test procedures; and



(3) procedures for emission testing.



2-  Apparatus and Reagents



     2.1  Measurement System.  The equipment required to extract,



transport, and analyze the gas sample constitutes the "measurement



system."  A schematic of the measurement system is shown in



Figure 20-1.  (Measurement system performance specifications  are



described in detail in Section 3.)  The essential components  of the



measurement system are described below.



     2.1.1  Probe.  Stainless steel type 316 or equivalent, to transport



gas from stack.
                               G-l

    -------
NJ
O
CD
O>
CD
3
Q.
O
Qj
(O
CD
CD
r-*
CD
                                   6-2

    -------
     2.1.2  Particulate Filter.   A filter fs used to remove particu-
lates ahead of the calibration valve assembly.   In most cases,
either in-stack or out-stack filter location is acceptable;  however,
out-stack filtration is required v/hen the sample gas temperature is
above 500°C (930°F).   The filtration temperature shall  be at least
120°C (250°F) to prevent moisture condensation.  Glass  fiber filters,
of the type specified in EPA Method 5, or equivalent, are recommended.
     2.1.3  Calibration Valve Assembly.  A three-way valve assembly is
used to direct the zero and span calibration gases to the analyzers.
This assembly shall be located directly behind  the probe and filter and
shall be capable of blocking the sample gas flow and introducing the
span and zero gases when the system is in the calibration mode.
     2.1.4  Calibration Gases.  Calibration gases are used to perform
zero, span and calibration checks of the analyzers during each test run.
The concentrations and specifications of these  gases are described in
detail in Sections 2.2 and 6.2.
     2.1.5  Heated Sample Line.   A FEP fluorocarbon or  stainless steel
(type 316 or equivalent) sample line is used to transport the gases to
the sample conditioner and analyzers.  The sample gas shall be maintained
at least 5°C (10°F) above the stack gas dew point to prevent moisture
condensation.
     2.1.6  Moisture Trap.  A moisture trap, designed to reduce the dew
point of the sample gas to 3°C (37°F) or less,  is used.  For instruments
not affected by water vapor, this device is not required; however, the
moisture content shall be determined using methods subject to the
approval of the Administrator and the NO  and CL concentrations shall
be corrected to a dry gas basis.
                              G-3

    -------
     2.1.7  Pump.   A nonreactive leak-free sample  pump is  used  to
pull the sample gas through the system at a flow rate sufficient
to minimize transport delay.   The pump shall  be  made from  or coated
with nonreactive material  (FEP fluorocarbon or type 316 stainless
steel).
     2.1.8  Sample Gas Manifold.  A sample gas manifold is pecommended
for diverting portions of the sample gas  stream  to the analyzers.  The
manifold may be constructed of glass, FEP fluorocarbon, or stainless
steel (type 316 or equivalent).  Instead  of using  the manifold,
separate sample lines may be connected to each analyzer.
     2.1.9  Oxygen Analyzer.   An oxygen analyzer is used to determine
the oxygen concentration (percent Op) of  the sample gas stream.
     2.1.10  Nitrogen Oxides Analyzer. A NO  analyzer is  used  to
                                            A
determine the ppm concentration of nitrogen oxides in the  sample gas
stream.
     2.1.11 Sulfur Dioxide Analysis.  Method 6 apparatus,  or equivalent,
is required for sulfur dioxide determination.
     2.2  Calibration Gas Specifications.
     2.2.1  Zero Gas.  Prepurified nitrogen is used.
     2.2.2  Nitrogen Oxide Calibration Gases. Mixtures of known con-
centrations of NO in nitrogen are required.  Nominal NO concentrations
of 25, 50, and 90 percent of the instrument full scale range are needed.
The 90 percent gas mixture is used to set and check the instrument span
and is referred to as span gas.  The 25 and 50 percent gas mixtures
shall be used to validate the analyzer calibration, prior  to each test.
                             G-4

    -------
     2.2.3  Oxygen Calibration Gases.   Ambient air at 20.9 percent
oxygen shall be used as the span gas (high range concentration gas).
A midscale calibration gas (approximately 13 percent Op in nitrogen)
shall be used to validate the analyzer calibration prior to each
test.
     2.2.4  Concentration Validation.   Within one month prior to test
use, calibration gases shall  be analyzed, by the appropriate test
method specified in Section 6.2, to determine their true concentration
levels.  Gas concentrations that are traceable to the National Bureau
of Standards and which can be demonstrated to be stable are exempted
from the analysis requirements.
3.  Measurement System Performance Specifications and Performance Test
      Procedures
     3.1  Analyzer.  "Span" is defined as the concentration range
(specified by manufacturer) over which an analyzer will give valid
readings.  The spans for the analyzers used in this method shall be
as follows:
     3.1.1  Oxygen Analyzer:   0 to 25% 02
     3.1.2  NO  Analyzer:  0 to 120 ppm
              X
     3.2  Analyzer Interferences and Interference Response.  The "inter-
ference response" of an analyzer is defined as the output response to
a component in the sample gas stream,  other than the gas component
being analyzed; the analyzers used in  this method shall not have a
total interference response of more than +2 percent of span.
                              G-5

    -------
     Participate matter and  water  vapor  are  the  primary interfering
species for most instrumental  analyzers, but these may be removed
physically by using filters  and  condensers.   Other possible  specific
interferents found in turbine  exhaust  streams  include carbon monoxide,
carbon dioxide, nitrogen oxides, sulfur  dioxide  and  hydrocarbons.   Each
analyzing instrument may respond to  one  or more  of these interferents
in ways that alter the desired measurement.
     The interference response of  an analyzer  is determined  by  measuring
the total analyzer response  to the gaseous components  (or mixtures)
listed in Table 20.1; these  gases  may  either be  introduced into the
analyzer separately, or as a single  gas  mixture. The total  interference
output response of the analyzer  to these components, if any, shall  be
determined (in concentration units).  The values obtained in an inter-
ference response test shall  be recorded  on a form similar to Figure  20.2.
If the sum of the interference responses of  the  test gases is greater
than 2 percent of the instrument span, the analyzer  shall not be used
in the measurement system of this  method.
     An interference response test of  each analyzer  shall be conducted
prior to its initial use in  the  field.  Thereafter,  if  changes  are
made in the instrumentation  which  could  alter the interference  response,
e.g., changes in the type of gas detector, the instruments shall be
retested.
     In lieu of conducting the interference  response test,  instrument
vendor data, which demonstrate that  for the  test gases  of Table 20.1
the interference performance specification  is not exceeded,  are
acceptable.  If these data are not available, the tests  shall  be made.
                              6-6

    -------
     TABLE 20.1   INTERFERENCE TEST GAS CONCENTRATIONS



                 CO                 500 ppm

                 S02       .         200 ppm


                 N0/N02             200 ppm

                 C02                10%

                 02                 20.9% (Air)
            FIGURE 20.2  INTERFERENCE RESPONSE
Date of Test:
Analyzer Type:	S/N_
Test Gas                      Analyzer Output
 Type            Cone.            Response           % of Span
       1 nf Snan  -  Analyzer Output Response     ,nn
       /0 of Span  ~      Instrument Span       x  10°
                             6-7

    -------
     3.3  Analyzer Response Time.   When a  change  in  pollutant  con-
centration occurs at the inlet of  the measurement system (i.e.,  at
probe), the chai.ge is not immediately registered  by  the  analyzer;
"response time" is defined as  the  amount of time  that  it takes for
the analyzer to register a concentration value  within  5  percent  of
the new inlet concentration.  The  maximum  response time  for the
^".ilyzers used in this method  is three minutes.
     To determine response time, first introduce  zero  gas into the
system until all readings are  stable; then, introduce  span gas into
the system.  The amount of time that it takes for the  analyzer to
register 95 percent of the final span gas  concentration  is the upscale
response time.  Next, reintroduce  zero gas into the  system; the  length
of time that it takes for the  analyzer output to  come  within 5 percent
of the final reading is the downscale response  time.   The upscale  and
downscale response times shall each be measured three  times.  The
readings shall be averaged, and the average upscale  or downscale response
time, whichever is greater, shall  be reported as  the "response time" for
the analyzer.  Response ,time data  are recorded  on a  form similar to
Figure 20.3.  A response time  test shall be conducted  prior to the
initial field use of the measurement system, and  shall be repeated  if
changes are made in the measurement system.
     3.4  Zero Drift.  "Zero drift" is the change in analyzer output
during a turbine performance test, when the input to the measurement
system is a pure grade of nitrogen (zero gas).  The  maximum allowable
                             6-8

    -------
                             RESPONSE TIfT
Dote of Tost
Analyser Type
Span Gas Concentre ticn_
Analyzer Spsn Setting_
Upscale
                    2
                    3
        Average upscale response
Dcv/riscalc1
                    2_
                    3
                                       S/N
                                    ppm
                                   ppm
                                   seconds
seconds
                                   seconds
                                           seconds
                                   seconds
seconds
                                   seconds
        Average dovnscale response
                                            seconds
Systen response time = slower average tiirr} =
                                                       seconds.
                              Fiqure 20.3
                                  G-9

    -------
zero drift for the analyzers  used  in  this  method  is  +2  percent  of  the           I
specified instrument span.  The zero  drift calculation  is  made  for
each gas for e? ,h turbine test run; this  is  done  by  taking the  difference
of the zero gas concentration values  measured  at  the start and  finish  of
the test (see Section 6.1).   The zero drift  is recorded (as a percentage
of the instrument span)  on  a  form  similar  to Figure  20.4.
     3.5  Span Drift.  "Span  drift" is the change in the analyzer
output during a turbine  performance test,  when the input to the measure-
ment system is span gas.   The maximum allowable span drift for  the
analyzers used in this method is +2 percent  of the specified instrument
span.  The span drift calculation  is  to be made for  each gas for each
turbine test run; this is done by  taking  the difference between the span
gas concentration values measured  at  the  beginning and  end of the  test.
Span drift is recorded (as  a  percentage of instrument span) on  a form
similar to Figure 20.4.   Span drift must  be  corrected for any zero
drift that occurred during  the test period (see Figure  20.4).                    •;
4.  Procedure for Field  Sampling
     4.1  Selection of a Sampling  Site and the Minimum  Number of Traverse
Points.
     4.1.1  Select a sampling site as close  as practical to the exhaust
of the turbine.  Turbine geometry, stack  configuration, internal
baffling, and point of introduction of dilution air  will vary for
different turbine designs.  Thus,  each of these factors must be given            |
special consideration in order to  obtain  a representative sample.
Whenever possible, the sampling site  shall be located upstream  of  the
                               6-10

    -------
                        TURCINE  SAMPLING SYSTFH
 Turbine: Type

 Date:
 Test No.:
 Analyzer:  Type
              Zero end Span Drift  Data

               	S/N
                                S/N
 Zero Gas
                        Initial
                     Calibration
                      ppm or  %
                               Final
                           Calibration
                            ppm  or %
  Difference
Initiel-Final
  ppm or %
of Snrn
 Mich Crililr.-ati
   r -.- ( ^n in
   V.'l.O V - I- — ••
% of span =
                                     of Dlfornce
                                                     x  100
                            Instrumant  Span
*Corrcctc:d for zero drift,  i.e.,  if zero drift over test period is +2 ppm
 then 2 1,'or.i shall bo subtracted  from the difference between the initial
 tncJ fir,:.! r;:adincs.
                                Fiaurs 20.4
                                    6-11

    -------
point of introduction of dilution air into the duct.   Sample  ports
may be located before or after the upturn elbow,  in order to  accom-
modate the configuration of the turning vanes  and baffles a^d to
permit a complete,  unobstructed traverse of the  stack.   The  sample
ports shall not be located v/ithin 5 feet or 2  diameters  (whichever  is
less) of the gas discharge to atmosphere.  For supplementary-fired,
combined-cycle plants, the sampling site shall be located between the
gas turbine and the boiler.
     4.1.2  The minimum diameter of the sample ports shall  be 3-inch
nominal pipe size (NPS).
     4.1.3  The minimum number of points for the  preliminary  Op
sampling (Section 8.3.2) shall be as follows:   (1) eight, for stacks
                                            •?          2
having cross-sectional areas less than 1.5 m1"  (16.1 ft ); (2) one
                           22                              2
sample point for each 0.2 m  (2.2 ft ) of area, for stacks of 1.5 m
         2                 2
to 10.0 m  (16.1 - 107.6 ft ) in cross-sectional  area; and (3) one
               :            2        2
sample point for each 0.4 m  (4.4 ft ) of area, for stacks greater
           2          2
than 10.0 m  (107.6 ft ) in cross-sectional area.  Note  that  for
circular ducts, the number of sample points must  be a multiple of 4,
and for rectangular ducts, the number of points must be  one of those
listed in Table 20.2; therefore, round off the number of points (upward),
when appropriate.
     4.2  Cross-sectional Layout and Location of  Traverse Points.  After
the number of traverse points for the preliminary Op sampling has been
determined, use Method 1 to locate the traverse points.
                              G-12

    -------
TABLE 20.2  CROSS-SECTIONAL  LAYOUT FOR  RECTANGULAR STACKS

     No.  of traverse               Matrix
         points                    layout
           9                       3x3
          12                       4x3
          16                       4x4
          20                       5x4
          25                       5x5
          30                       6x5
          36                       6x6
          42                       7x6
          49                       7x7
                            6-13

    -------
     4.3  Measurement System Operation.
     4.3.1  Preliminaries.
     4.3.1.1   rn'or to the  turbine  test,  the measurement  system  shall
have been demonstrated to have met  the performance  specifications  for
interference  response and response  time described in  Sections  3.2  and
3.3.
     4.3.1.2   Turn on the sample pump  and instruments;  allow  the normal
warnup time required for stable instrument operation.
     4.3.1.3   After the instruments have  stabilized,  the  measurement
system shall  be calibrated  using the procedures  detailed  in Section 6.1.
Transfer the  zero and span  gas calibration data  from  Figure 20.5 to a
form similar  to Figure 20.4.
     4.3.1.4   At the beginning of each NO  test  run and,  as applicable,
                                         /\
during the run, record turbine data as indicated in Figure 20.6   Also,
record the location and number of the  traverse points on  a diagram.
     4.3.2  Preliminary Oxygen Sampling.
     4.3.2.1   At the start  of a 3-run  sample  sequence,  position  the
probe at the  first traverse point and  begin sampling.   The minimum
sampling time at each point shall be 1 minute  plus  the  average system
response time.  Determine the average  steady-state  concentration of
Op at each point and record the data on Figure 20.7.
     4.3.2.2   Select the-eight sample  points  at  which the lowest qxygen
concentrations were obtained.  These same points shall  be used for
all three runs which comprise the emission test. More  points may be
used, if desired.
                              6-14

    -------
                              Figure  20.5



                                     ;*1  OAT A
Date
Analyzer  lype
High Range Gas Conc.
Mid Ranee Gas Cone.
Low Range Gas Cone.
Zero Gas
S/IJ
% Full Scale_




% Full Scale_




% Full Scale_




% Full Scale
                                  6-15

    -------
                            Figure  20.6

                       STATIONARY GAS  TURBINE
Tost Operator
Turo nc ID
Location
             Type_

             S/N
            Plant
            City_
Ambient Temperature_


Aiiibi en t Hurni di ty	


Test Time Start
Test Tims Finish_


Fuel Flow Rate
                         DifJE OPERATIC;]  RECORD
                                         Date
Ultimate Fuel
  AnalysisC
                                                     H
                                                     0
                                                      Ash
                                                      HoO
                                         Trace Metals
                                                    Na
                                                     Va
                                                     etc.
                                         Operating Load_
 Hater or Steam	
   Flow Rate

 Ambient Pressure 	
 Describe? i, jGsureiiiiint melliod,  i.e.,  continuous  flow meter,
  start finish volumes, etc.

**i.c., Additional elements added for  smoke  suppression.
                                  G-16

    -------
                          FIGURE 20.7
                 Preliminary Oxygen Traverse
Location
Date
     Plant
     City, State
Turbine ID
     Mfg.	
     Model, serial  number ^	
     Sample Point            Oxygen Concentration
                          G-17

    -------
     4.3.3  Emission Sampling.



     4.3.3.1  Position the probe at the first point determined in



the preceding section and begin sampling.   The minimum sampling time



at each point shall be 3 minutes plus the  average system response



time.  Determine the average steady-state  concentration of CL apd NO
                                                            f*      "A


at each point and record the data on Figure 20.8.



     4.3.2.2  After sampling the last point, conclude the test run by



recording the final turbine operating parameters and by determining



the zero and span drift, as described in Sections 3.4 and 3.5.  If



the zero and/or span drift exceed +2.0 percent the run may be



considered invalid, or may be accepted provided the calibration dat9



which results in the highest corrected emission concentration is used.



     4.3.2.3  If additional turbine runs are conducted within 4 hours



of the previous run, an initial calibration of the measurement system



is not required.  If more than 4 hours have elapsed between runs, the



pretest calibration shall be done.



     4.4  An SOp determination shall be made (using Method 6, or



equivalent) during the test.  A minimum of six total points, selected



from those required for the NO  measurement, shall be sampled; two
                              A


points shall be used for each sample run.   The sample time at each



point shall be at least 10 minutes.  The oxygen readings taken during



the NO  test runs corresponding to the S0? traverse points (see
      y\                                  ^*


Section 4.3.3.1) shall be averaged, and this average oxygen concen-



tration shall be used to correct the integrated SOp concentration



obtained by Method 6 to 15 percent Op (see Equation 20-1).
                                 G-18

    -------
                                  00
                                        oo.
                             CU     >
                             S.
                             O
                            -P
                             _( LJ_ J_
' **1
>- LU °
or _J .
$5: S
^•~ -K
O «aC ?
•— ' to _
h- °
«=C oo
i— <:
00 C3
1C ***^











z:





oo
o3
r— 4J













QJ







'cu c: >,' 43
Cn "O fO 4J n3
4- O •— ••- 4-)
21 S D- O 00
OL







C/>
CO







s
•f—
-a
•r-








E CO E
Q CU i_ 3
»— i 1 — a. on
CU O -M 4J 4->
c ••- c c c
•i- -4J CU OJ CJ
TTf_j rO »f" *^* •f" UJ -4"^ «H*







4->
J-
ra
-(-) -f—







.G
l/l
C
• |»V
X E
O Q.
Z Q.
CM fe^
O 	
CD C
£ £
h- ^-^

CU
r— 4->
a. c
E -r-
rO O
OO Q-
CO U.
P




0



-


OJ
J-
E
e
if-
CJ
rs
r- *
OJ
4->
3
v>
•^
s
•M
to
ai
en

?

•r- •f—
1 	 1 	
I — I —

• \ • *
s- u j3ja^3+->toto
3 O E 6 E (0 fi "'
                                                      Q-19

    -------
5.  Emission Calculations



     5.1  Correction to 15 Percent Oxygen.   Using Equation 20-1,



calculate the NO  and S09 concentrations (adjusted to 15 percent 09).
                X       c                                         c.


The correction to 15 percent oxygen is sensitive to the accuracy of



the oxygen measurement.  At the level  of analyzer drift specified in



the method (+2 percent of full  scale), the  change in the oxygen concen-



tration correction can exceed 10 percent when the oxygen content of



the exhaust is above 16 percent 0^.  Therefore Op analyzer stability



and careful calibration are necessary.




     Actual Pollutant         j- Q,,

      Concentration  x  on- Qy—VV ;v*; ',1   =  Pollutant concentration

      (NOX or S02)      20'9^ ' V° actua1       adjusted to 15% Og




                                                Equation 20-1



where:



     5.9% is 20.9% - 15% (the defined concentration basis)



     Op actual is the sample point oxygen concentration for NO
      t.                                                       J\


       calculation, and the average Op concentration for S0? calculation.



     5.2  Calculate the average adjusted NO  concentration by summing
                                           A


the  point values and dividing by the number of sample points.



6.  Calibration



     6.1  Measurement System.  Prior to each turbine test, the measurement



system shall be calibrated according to the procedures described below.



The manufacturer's operation and calibration instructions are also to



be followed as required for each specific analyzer.
                             G-20

    -------
     6.1.1  Turn on all measurement system "components and allow them



to warm up until stable conditions are achieved.  fJext, introduce



zero gas and each of the calibration gases described in Section 6.2,



one at a time, into the inlet of the probe.   The responses of the



analyzer to these gases shall be used to establish a calibration curve



or to verify the manufacturers calibration curve.  The data obtained



in these procedures shall  be recorded on a form similar to Figure 20.4.



If, for the mid-scale gases, the accuracy of the manufacturer's



calibration curve or the expected response curve cannot be shown to



be +2 percent of full scale (or better), the calibration shall  be



considered invalid and corrective measures on the instrument shall  be



taken.  The calibration procedure shall  be repeated, using only zero



gas and span gas, at the conclusion of test; this allows calculation



of zero and span drift (Sections 3.2 and 3.3).



     6.2  Calibration Gas  Mixtures.



     6.2.1  Within one month prior to the turbine test, the NO  calibra-
                                                              A


tion gas mixtures shall be analyzed, using the  phenoldisulfonic acid



procedure (Method 7) for nitrogen oxides.  A minimum of three analyses



shall be done, and the average concentration of each gas shall  be



reported as the true calibration gas value (see Figure 20.9).  Alternate



procedures may be employed, subject to the approval  of the Administrator,



to determine the calibration gas concentration.



     Note:  The NO  calibration gas mixtures shall contain nitric oxide
                  /\


(NO) in nitrogen.  Instruments which require conversion of one  nitrogen
                              G-21

    -------
                              Figure 20.9



                ANALYSIS  OF CALIUMl IG.'i GAS MIXTURES
CYLINiO GAS  LOIVOSITION                   Reference  Method Used
          Lav__R.'n_no_ Cajjbnition Cos Mixture



          Sample  1	ppm



          Sample  2	ppin



          Sample  3	ppm



          Average	ppm





          Mid  Range C a lib rah'on Gas Mixture



          Sample  1	ppm



          Sample  2	ppm



          Sample  3	ppm



          Avcrage_	ppm





          HIn'n Rpnnp (r.ppn) Gal 1i;ra1.1oii  Gas  Mi_x1._nro



          Sample  1	ppm



          Sample  ?.	ppm



          Sample  3	ppm



          A v c r c? cj o	p pm
                                   G-22

    -------
oxide component to another for total  NO  measurement shall  be
                                       A


checked to ensure that this conversion is complete and reproducible,



as specified by the manufacturer.



     6.2.2  Ambient air may be used as the oxygen span gas.   The



mid-scale calibration gas concentration shall  be certified  (by vendor)



as being within +2 percent of the indicated concentration.
                              G-23

    -------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1  REPORT NO.
                                                           3. RECIPIENT'S ACCESSION-NO
4 TITLE AND SUBTITLE
 Standards Support and Environmental  Impact Statement,
 Volume 1:  Proposed Standards  of Performance for
 Stationary Gas Turbines	
                                                         5. REPORT DATE
                                                            September  1977
                                                         6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
 Standards Development Branch
 Emission otandards and Engineering Division
 Research Triangle Park, N.  C.   27711
                                                           10. PROGRAM ELEMENT NO.
                                                         11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
 DAA for Air Quality Planning  and Standards
 Office of Air and Waste  Management
 U.S.  Environmental Protection Agency
 Research Triangle Park,  N.  C.  27711
                                                           13. TYPE OF REPORT AND PERIOD COVERED
                                                         14. SPONSORING AGENCY CODE

                                                            EPA/200/04
15. SUPPLEMENTARY NOTES
                      Volume 1  discusses the proposed  standards and the resulting
 environmental and  economic effects.  Volume 2,  to  be published when the  standards are
 promulgated, will  contain a summary of public comments,  EPAresponse^and  a  discussion
16.ABSTRACT    Of differences between the proposed and promulgated standards.
 Standards of performance  to control nitrogen oxides  and sulfur dioxide emissions
 from new, reconstructed and modified stationary gas  turbines in the U.S.  are  being
 proposed under section 111  of the Clean Air Act.  This  document contains  information
 on the gas turbine  industry and emission control  technology, a discussion of  the
 selected emission limitations and the supporting  data and the alternatives which
 were considered, and  analyses of the environmental  and  economic impacts of the
 proposed standards.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                                                                       c.  COSATI Field/Group
  Gas turbines
  Air pollution  control  equipment
  Standards of performance
  Pollution control
                                             Air  pollution control
                                             Sulfur dioxide
                                             Nitrogen oxides
                                             Water injection
18.
DISTRIBUTION STATEMENT
Unlimited.  Available from Public Informa-
tion Center (PM-215), Environmental
Protection Agency, Washington, D.C. 20460
                                              19. SECURITY CLASS /ThisReport)
                                                Unclassified
                                                                         21. NO. OF PAGES
                                              20. SECURITY CLASS (Thispage)
                                                Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (9-73)

    -------