NTG

TW-0076
03 May 2002
NEXT GENERATION TURBINE PROGRAM
TECHNICAL PROGRESS REPORT

FINAL REPORT
PERIOD OF PERFORMANCE: 16 AUGUST 2000 TO 14 JUNE 2002
Prepared for
National Energy Technology Center
U.S. Department of Energy
AAD Document Control M/S 921-107
P.O. Box 10940
Pittsburgh, PA 15236
Prepared under
Contract DE-AC26-00NT40847
Prepared by
Pratt & Whitney
Advanced Engine Programs
400 Main Street
East Hartford, CT 06108
DISCLAMER NOTICE
THIS REPORT WAS PREPARED AS AN ACCOUNT OF WORK SPONSORED BY AN AGENCY OF THE U.S.
GOVERNMENT. NEITHER THE U.S GOVERNMENT NOR ANY AGENCY THEREOF, NOR ANY OF THEIR
EMPLOYEES, MAKES ANY WARRANTY, EXPRESS OF IMPLIED, OR ASSUMES ANY LEGAL LIABILITY OR
RESPONSIBILITY FOR THE ACCURACY, COMPLETENESS, OR USEFULNESS OF ANY INFORMATION,
APPARATUS, PRODUCT, OR PROCESS DISCLOSED, OR REPRESENTS THAT ITS USE WOULD NOT
INFRINGE PRIVATELY-OWNED RIGHTS. REFERENCE HEREIN TO ANY SPECIFIC COMMERCIAL PRODUCT,
PROCESS, OR SERVICE BY TRADE NAME, TRADEMARK, MANUFACTURER, OR OTHERWISE DOES NOT
NECESSARILY CONSTITUTE OR IMPLY ITS ENDORSEMENT, RECOMMENDATION, OR FAVORING BY THE
U.S. GOVERNMENT OR ANY AGENCY THEREOF. THE VIEWS AND OPINIONS OF AUTHORS EXPRESSED
HEREIN DO NOT NECESSARILY STATE OR REFLECT THOSE OF THE U.S. GOVERNMENT OR ANY AGENCY
THEREOF.
Published by Business Development Technical Writing

Pratt & Whitney TW-0076
ABSTRACT
The Next Generation Turbine (NGT) Program’s technological development focused on a study of the feasibil-
ity of turbine systems greater than 30 MW that offer improvement over the 1999 state-of-the-art systems. This pro-
gram targeted goals of 50 percent turndown ratios, 15 percent reduction in generation cost/kW hour, improved ser-
vice life, reduced emissions, 400 starts/year with 10 minutes to full load, and multiple fuel usage. Improvement in
reliability, availability, and maintainability (RAM), while reducing operations, maintenance, and capital costs by
15 percent, was pursued. This program builds on the extensive low emissions stationary gas turbine work being
carried out by Pratt & Whitney (P&W) for P&W Power Systems (PWPS), which is a company under the auspices
of the United Technologies Corporation (UTC). This study was part of the overall Department of Energy (DOE)
NGT Program that extends out to the year 2008. A follow-on plan for further full-scale component hardware testing
is conceptualized for years 2002 through 2008 to insure a smooth and efficient transition to the marketplace for
advanced turbine design and cycle technology. This program teamed the National Energy Technology Laboratory
(NETL), P&W, United Technologies Research Center (UTRC), kraftWork Systems Inc., a subcontractor on-site at
UTRC, and Multiphase Power and Processing Technologies (MPPT), an off-site subcontractor.
Under the auspices of the NGT Program, a series of analyses were performed to identify the NGT engine sys-
tem’s ability to serve multiple uses. The majority were in conjunction with a coal-fired plant, or used coal as the
system fuel. Identified also was the ability of the NGT system to serve as the basis of an advanced performance
cycle: the humid air turbine (HAT) cycle. The HAT cycle is also used with coal gasification in an integrated cycle
HAT (IGHAT). The NGT systems identified were:
• Feedwater heating retrofit to an existing coal-fired steam plant, which could supply both heat and peaking
power (Block 2 engine)
• Repowering of an older coal-fired plant (Block 2 engine)
• Gas-fired HAT cycle (Block 1 and 2 engines)
• Integrated gasification HAT (Block 1 and 2 engines).
Also under Phase I of the NGT Program, a conceptual design of the combustion system has been completed.
An integrated approach to cycle optimization for improved combustor turndown capability has been employed.
The configuration selected has the potential for achieving single digit NOx/CO emissions between 40 percent and
100 percent load conditions. A technology maturation plan for the combustion system has been proposed.
Also, as a result of Phase I, ceramic vane technology will be incorporated into NGT designs and will require
less cooling flow than conventional metallic vanes, thereby improving engine efficiency. A common 50 Hz and
60 Hz power turbine was selected due to the cost savings from eliminating a gearbox. A list of ceramic vane tech-
nologies has been identified for which the funding comes from DOE, NASA, the U.S. Air Force, and P&W.
ii
ACKNOWLEDGEMENTS
The Program Manager for the DOE was Kanwal Mahajan, from the beginning of the contract in July 2000 until
July 2001. Kanwal moved to a different job within DOE and was replaced by Chuck Alsup.
On the contractor side, the work involved a great many people in different component groups. The leaders of
the major efforts are as follows; all are from P&W except as noted.
Program Manager – Bill Day
Performance – Jill Kelly
Compression System – Norm Jacques
Combustion System – Steve Morford
Combustion Modeling – Paul Van Slooten (UTRC)
Ceramic Materials – Gary Linsey and Ellen Sun (UTRC)
Turbine Systems – Rajendra Agrawal and Hector Pinero
Mechanical Systems – Wes Surowka
Rotor Dynamics – Lisa Marquis
Systems Integration – Mike Babu
Balance of Plant – Brian Donnelly
Bottoming Cycle – Lance Hays Douglas Energy)
Bottoming Cycle – Ted Bond (MPPT)
Market Study – Stan Blazewicz and Dinesh Agarwal (Arthur D. Little Company [ADL])
Advanced Cycles – Fred Robson (kraftWork Systems)
Program Administration – Marc Lamoureux
Contracts – Leigh Fournier.
iii
CONTENTS
Section Page
ABSTRACT ........................................................................................................................................................ ii
ACKNOWLEDGEMENTS................................................................................................................................ iii
EXECUTIVE SUMMARY ..................................................................................................................................x
1. EXPERIMENTAL............................................................................................................................................ 1-1
1.1 Introduction ........................................................................................................................................ 1-1

1.2 Overall System Description and Attributes ........................................................................................ 1-1
1.2.1 Performance ........................................................................................................................... 1-2
1.2.2 Emissions ............................................................................................................................... 1-2
1.3 Economics ......................................................................................................................................... 1-2

1.3.1 Capital Cost ........................................................................................................................... 1-2
1.3.2 Operating and Maintenance Cost: ......................................................................................... 1-2
1.3.3 Economic Analysis ................................................................................................................ 1-3
1.4 Reliability, Availability, and Maintainability..................................................................................... 1-4
2. RESULTS AND DISCUSSION....................................................................................................................... 2-1
2.1 Component System Descriptions....................................................................................................... 2-1

2.1.1 Compression System ............................................................................................................. 2-1
2.1.2 Combustion System ............................................................................................................... 2-3
2.1.3 Turbine System .................................................................................................................... 2-13
2.1.4 Ceramic Turbine Vanes and Ceramic Matrix Composite Liners ........................................ 2-16
2.1.5 Mechanical Systems ............................................................................................................ 2-18
2.1.6 Control System .................................................................................................................... 2-22
2.2 Combined Cycle ............................................................................................................................... 2-26
3. CONCLUSION................................................................................................................................................. 3-1
3.1 Market Studies .................................................................................................................................... 3-1
3.2 Public Benefits.................................................................................................................................. 3-10

3.2.1 The System’s Ability to Serve Multiple Uses ..................................................................... 3-10
3.2.2 Support of Vision 21 ........................................................................................................... 3-19
3.2.3 Next Generation Turbine/Fuel Cell Hybrids ....................................................................... 3-19
3.2.4 Comments on Public Benefits of the Vision 21 Application .............................................. 3-20
4. REFERENCES ................................................................................................................................................. 4-1
iv
FIGURES
Figure Page
1-1. Intercooler Cycle........................................................................................................................................ 1-1
1-4. Combined Cycles Cost Considerably More Than Simple Cycles ............................................................. 1-2
1-2. FT60 Provides Gamechanger Performance ............................................................................................... 1-2
1-3. Market Price Trends Lower With Larger Size........................................................................................... 1-2
1-5. NGT Is Competitive Over a Wide Range of Operating Hours .................................................................. 1-3
1-6. Comparison of the Relative Attractiveness of NGTs By Calculating the NPV
of Production Cost/kW for Each Unit Over a 15-Year Investment Horizon ............................................ 1-4
1-7. Relative Competitiveness Between NGT and FX2.................................................................................... 1-4
1-8. Aeroderivatives With Lease Engines Have High Reliability and Availability.......................................... 1-4
2-1. FT60 Compression System ........................................................................................................................ 2-1
2-2. Low-Pressure Compressor Module............................................................................................................ 2-2
2-3. High-Pressure Compressor......................................................................................................................... 2-3
2-4. Typical NOx Emissions Versus Flame Temperature ................................................................................. 2-4
2-5. Conceptual Combustor Design .................................................................................................................. 2-4
2-6. Effect of Compressor Inlet Guide Vane Modulation and Low-Compressor Bleed
On Combustor Exit Temperature Versus Load.......................................................................................... 2-5
2-7. Flame Temperature Distribution Versus Load........................................................................................... 2-5
2-8. Schedule for Development Activities ........................................................................................................ 2-6
2-9. Computational Cost Estimate for a Hybrid Finite Volume/Full Composition-PDF Method .................... 2-8
2-10. Particle Tracking Sketch .......................................................................................................................... 2-11
2-11. CFD Model and Grid ............................................................................................................................... 2-12
2-12. CFD Calculation Mixing Profile.............................................................................................................. 2-12
2-13. Convergence of Exit Mass Flow Rates .................................................................................................... 2-12
2-14. Convergence of Mean Mixture Fraction at Selected Vertex.................................................................... 2-12
2-15. Convergence of Mixture Fraction Variance at Selected Vertex .............................................................. 2-13
2-16. Composition-PDF Mixing........................................................................................................................ 2-13
2-17. Timeline and Tasks for Block 1 FT60IC ................................................................................................. 2-14
2-18. FT60IC Turbine Cross-Section ................................................................................................................ 2-15
2-19. FT60 IC Combustor and Portion
of the Turbine........................................................................................................................................... 2-16
2-20. Cooling Flow............................................................................................................................................ 2-17
2-1. Increased Revenue With Ceramic and CMC Components ...................................................................... 2-18
2-21. Isometric View of the FT60 IC With a Water-Cooled Intercooler .......................................................... 2-25
v
Figure Page
2-22. Plan View of the FT60 IC With a Water-Cooled Intercooler .................................................................. 2-25
2-23. Isometric View of the FT60 IC With an Air-Cooled Intercooler............................................................. 2-25
2-24. Plan View of the FT60 IC With an Air-Cooled Intercooler..................................................................... 2-26
2-25. Biphase Combined-Cycle Boasts System Efficiency .............................................................................. 2-27
2-26. Heat Transfer for the Biphase Bottoming Cycle...................................................................................... 2-28
3-1. Framework of the Scenarios for the ADL Analysis................................................................................... 3-2
3-2. Comparative Economic Analysis............................................................................................................... 3-2
3-3. Market for New Generating Capacity in the U.S....................................................................................... 3-3
3-4. Generation Additions in the U.S. ............................................................................................................... 3-3
3-5. NGT Sales Forecasts .................................................................................................................................. 3-4
3-6. Benefits of Adopting NGTs ....................................................................................................................... 3-4
3-7. Savings Were Generated Using Market Projections .................................................................................. 3-5
3-8. Capacity of NGTs ...................................................................................................................................... 3-5
3-9. U.S. Fuel Price Projections ........................................................................................................................ 3-5
3-10. Efficiencies, Emission Factors, and Displaced Generation Technologies................................................. 3-6
3-11. Fuel Savings ............................................................................................................................................... 3-6
3-12. Fuel Cost Reduction................................................................................................................................... 3-7
3-13. Reduction in CO2 Emissions...................................................................................................................... 3-7
3-14. Reduction of Other Emissions ................................................................................................................... 3-8
3-15. Cumulative Energy and Emissions Savings............................................................................................... 3-8
3-16. Advantages of NGTs Over Steam Plants ................................................................................................... 3-9
3-17. Projected Sales of NGTs Worldwide ......................................................................................................... 3-9
3-18. Typical Daily Load Factor for Feedwater Heating Plant ......................................................................... 3-10
3-19. Diagram of Feedwater Heating Cycle...................................................................................................... 3-11
3-20. Base-Load Supercritical Steam Plant....................................................................................................... 3-11
3-21. Feedwater Heating for Peak Power.......................................................................................................... 3-12
3-22. Schematic of Steam System ..................................................................................................................... 3-13
3-23. Schematic of Repowered Plant ................................................................................................................ 3-13
3-24. Simplified Diagram of HAT Cycle .......................................................................................................... 3-15
3-25. HIPPS With NGT-Type Gas Turbine ...................................................................................................... 3-16
3-26. A HIPPS/HAT System............................................................................................................................. 3-17
3-27. IGHAT Schematic.................................................................................................................................... 3-18
3-28. NGT/SOFC Hybrid .................................................................................................................................. 3-19
vi
TABLES
Table Page
2-1. NGT Program Turbine Technologies....................................................................................................... 2-15
2-2. Data Comparing the Pressure Drops Data Calculated for These Two Header Options........................... 2-25
3-1. NGT Market Potential in U.S. (2007 to 2020)........................................................................................... 3-3
3-2. NGT Market Potential in the Rest of World (2005 to 2020) ..................................................................... 3-4
3-3. FT60 IGHAT Performance Estimates (Texaco HEQ Gasifier) ............................................................... 3-18
vii
Acronyms
A
ADL Arthur D. Little Company

ATS Advanced Turbine Systems
C
CFD Computational Fluid Dynamics

CMC Ceramic Matrix Composite
COE Cost of Electricity
D
DLE Dry Low Emissions

DOE Department of Energy
E
EBC Environmental Barrier Coating

H
HAT Humid Air Turbine

HEQ High-Efficiency Quench
HHV High Heat Value
HIPPS High-Performance Power Systems
HPC High-Pressure Compressor
HPT High-Pressure Turbine
HRSG Heat Recovery Steam Generator
I
IGCC Integrated Gasification Combined Cycle

IGHAT Integrated Humid Air Cycle
J
JIC Jet-In-Crossflow
L
LES Large Eddy Simulations

LFL Lower Flammable Level
LHV Lower Heating Value
LPC Low-Pressure Compressor
LPS Low-Pressure Shaft
LPT Low-Pressure Turbine
viii
Acronyms
M
MPPT Multiphase Power and Processing Technologies

N
NETL National Energy Technology Laboratory

NGT Next Generation Turbine
NPV Net Present Value
O
O&M Operating and Maintenance

P
P&W Pratt & Whitney

PWPS Pratt & Whitney Power Systems
R
RAM Reliability, Availability, and Maintainability

S
SOFC Solid Oxide Fuel Cells

U
UTC United Technologies Corporation

UTRC United Technologies Research Center
ix
EXECUTIVE SUMMARY
This report is a summary of work conducted by P&W on behalf of the DOE’s NETL. For the NGT Program,
P&W performed a feasibility and program planning study of turbine systems greater than 30 MW in preparation for
the development of an advanced gas turbine system that would allow for improvement over 1999 state-of-the-art
systems while meeting the following objectives:
• Increase lower heating value net system efficiency by 15 percent or more
• Improve turndown ratios (using a turbine at partial capacity) by 50 percent or more
• Improve service life
• Reduce emissions of carbon and nitrogen oxide gases
• Reduce operations, maintenance, and capital costs by 15 percent or more
• Offer flexibility for at least 400 starts per year and 10 min to full load from the start of command
• Improve RAM
• Provide the capability to use multiple fuels.
An intercooled cycle was selected as the basis for the study. Conceptual design work was done on this cycle in
enough detail to determine that it was feasible and would meet the objectives of the program. With an output of 140
to 170 MW at low-capital cost, and an efficiency of 49 to 50 percent, this product has considerable advantages over
existing systems in a wide range of peaking and intermediate load applications. This was borne out by the market
study and economic analysis conducted by ADL.
The intercooled cycle can also serve multiple uses involving coal. Using the heat from the intercooler and tur-
bine exhaust for feedwater preheating the intercooled cycle improves the efficiency of new or existing coal-fired
plants and enables the owner to sell power in on-peak, as well as base load, markets. This enhances the economic
competitiveness of conventional coal-fired plants.
Using the same compressors and intercooler, the intercooled cycle can be converted to a HAT cycle configura-
tion; the combustor and turbine would be enlarged to accommodate the additional moisture. Integrated with a coal
gasifier, the HAT cycle would produce efficiencies as high as an integrated gasification combined cycle (IGCC),
but at reduced capital cost, thereby addressing an issue with IGCC that has slowed market penetration in coal-
based applications. The moisture in the HAT cycle retards emissions, as shown in tests at DOE facilities under the
HAT Cycle Technology Development Program that is part of the Advanced Turbine Systems (ATS) Program.1 The
moisture could also enable a lean premix combustor for the HAT cycle in a coal gasification cycle, which is not
feasible with a conventional IGCC due to the hydrogen content in the coal gas. This would further reduce emis-
sions from the coal gasification-based cycle.
Longer term, the HAT cycle and coal gasifier can be integrated with fuel cells in a hybrid cycle to meet the 60
percent efficiency goal of the Vision 21 Program.
Technological developments required for the intercooled cycle are in the areas of high-pressure combustion,
ceramic materials, advanced alloys and coatings, and intercooler and compressor integration. Extending the inter-
cooled cycle to the HAT cycle integrated with a coal gasifier requires combustion development for the high mois-
ture and coal gas fuel, and development testing of turbine materials and coatings to accommodate the high moisture
content.
1 DOE contract number DE-AC21-96MC33084
viii
1. EXPERIMENTAL
1.1 INTRODUCTION
The NGT Program focuses on turbine systems greater
than 30 MW that improve on the 1999 state-of-the-art sys- Intercooler
tems. Program goals include a 15 percent improvement in

Combustor
efficiency, a 15 percent reduction in operating, mainte-

nance, and capital cost, a reduction in generation cost/kW
HPC HPT
hour, improved service life, reduced emissions, 400 starts/ LPC LPT
year with 10 min to full load with multiple fuel usage, and
a 50 percent improvement in turndown ratios. Improve- • Twice the output of simple cycle - dollars per kiloWatt
• Efficiency 45-50% - Competitive at midrange capacity factor
ment is also sought in RAM. The current contract con- • Rapid start to full load - Quick dispatch to meet fluctuating
loads with multiple daily on/off cycles
cerns Phase 1, feasibility study and program planning.
126727.cdr
The technical approach chosen to meet the goals of
Figure 1-1. Intercooler Cycle
the NGT Program is to use the intercooled cycle, depicted
schematically in Figure 1-1.
The initial direction of the NGT Program was to use an aircraft engine high-pressure compressor (HPC) as the
basis for the industrial gas turbine. This would have resulted in an intercooled cycle in the 60-MW output class.
The NGT Program direction changed following a program review with PWPS technical management in December
2000 and subsequent discussions with P2 Energy, the sales and marketing arm of PWPS. The decision was made to
increase the size of the industrial gas turbine to over 100 MW in order to meet what is believed to be the market
demand. This demand was confirmed by the market survey carried out by ADL and P2 Energy. The resulting sys-
tem, called the FT60 IC, offers considerable public benefit in providing very high efficiency at low-capital cost in
an output class that meets the needs of the major power generators for peaking and intermediate load power.
1.2 OVERALL SYSTEM DESCRIPTION AND ATTRIBUTES

Figure 1-1 features a low-pressure shaft (LPS) driving the generator directly at 3,600 rpm. The HPC will be
designed only for industrial gas turbine requirements. This makes the job considerably lower risk than developing
an HPC that must also meet aircraft engine requirements of maneuver loads, sudden accelerations, light weight and
HPC discharge temperature. The HPC case can be made compatible with the combustor case, which must be
heavier than that of an aircraft engine due to the large cans required for dry low emissions (DLE). Thus, the back-
bone of the machine can be made strong enough to support the combustor, without compromising an aircraft
engine HPC.
The LPS has relatively low inertia, since the low-pressure compressor (LPC) has only four stages; therefore,
the starting motor can accelerate the LPS relatively quickly to achieve fast starts.
The combustors are individual cans, each with its own convectively-cooled transition piece connecting to the
1st-stage vanes.
The FT60 IC will be developed in phases. In the Block 1 configuration, the combustor exit temperature is in
the 1316°C (2400°F) class with conventional materials and coatings. The airflow is about 224 kg/sec (600 lb/sec)
and the overall pressure ratio is in the 50 to 55 range. The LPC discharge of the FT60 IC is about 0.304 Mpa (3
atm). The exhaust temperature is below 371°C (700°F).
The Block 2 configuration uses the same compression system as in Block 1, but represents an uprate of the hot
section to a combustor exit temperature in the 1482°C (2700°F) class and the introduction of ceramic vanes and
advanced metallic blade materials with advanced thermal barrier coatings. The exhaust temperature is 438°C
(820°F) to 449°C (840°F).
1-1
1.2.1 Performance
From Figure 1-2 it can be seen that the efficiency is about 55
49 percent for Block 1 and 50 percent for Block 2. The output 50
increases considerably from Block 1 to Block 2, from about FT60 IC Block 1 FT60 IC Block 2
Efficiency
140 MW to about 170 MW. This reduces the dollars/kW cost 45
Trent
after accounting for increased materials cost. 40 LM6000 7FB
FT8-3 FT8-3 Twinpac GT24
FT8 GTX100 M501F
Compared to 1999 state-of-the-art systems, as reported in 35 6FA 7EC 7FA
7EA
the Gas Turbine World 1999-2000 Handbook, the efficiency 30
of the FT60 IC Block 2 is more than 20 percent higher than 0 50 100 150 200
MW
the most efficient aeroderivatives and 33 percent higher then 126728.cdr
the most efficient frame-type gas turbines. Figure 1-2. FT60 Provides Gamechanger
1.2.2 Emissions Performance
The objective is to achieve 9 ppm NOx emissions after an
introductory period at 25 ppm NOx emissions, on gas fuel, and 25 ppm NOx emissions with water injection on liq-
uid fuel. The very low exhaust temperature of the FT60 IC provides an advantage in meeting stringent emissions
requirements at reasonable cost, because it accommodates the use of a conventional selective catalytic reactor with
no need to reduce the exhaust temperature by outside air dilution or a bottoming cycle; this enables NOx emissions
of less than 3 ppm. This is particularly important for the FT60 IC, since its high efficiency will drive the economic
dispatch point well into midrange capacity factor for many installations, where local emissions caps in tons per
year often require extremely low NOx.
1.3 ECONOMICS
1.3.1 Capital Cost
From a capital cost standpoint, the target market price 425 FT8 DLE
(cost to the end user) is in the same range as that of frame- FT8
375 GT10B LM6000 PD
type gas turbines (for Gas Turbine World scope, including LM2500 PK
Dollars/KW
FT8 Twin
the intercooler, see Figure 1-3). This price range is consider- 325 LM6000 PC Target Price Range for FT60 IC
GTX100 Trent DLE
ably lower then that of combined cycles (Figure 1-4). 275
LM6000 Sprint
GT8C2
6 FA
V64.3A
1.3.2 Operating and Maintenance Cost: 225 7EA

GT11N2 7FA
The largest single piece of operating and maintenance W501D5A
W501F
175
(O&M) cost is typically the repair and replacement of turbine 0 25 50 75 100 125 150 175 200
MW
airfoils that amounts to about half the total. Compared to 126729.cdr
large frame-type gas turbines, the FT60 IC has differences in
Figure 1-3. Market Price Trends Lower With
the design that favorably affect O&M cost.
Larger Size
The FT60 IC is being designed for a hot gaspath inspec- 800
tion at 25,000 hr and a major inspection at 50,000 hr. The V64.3A 6 FA
700
high-pressure turbine (HPT) airfoils will require coating LM6000 PC
2 X LM6000
Trent DLE
Dollars/KW
2 X 1 Trent
refurbishment at 25,000 hr and replacement at 50,000 hr. 600 FT8 Twin
Cyclic life will be at least 5,000 cycles. For the low-pressure 7EA
W501D5A
500
turbine (LPT), airfoil coating refurbishment will be required 2 X 6FA
7FA
at 50,000 hr, where coatings are present and the design life is 400 W501F
2 X 7FA
100,000 hr. Cyclic life is 15,000 cycles for LPT airfoils. 300
2 X 501F
Based on experience with the FT4 and FT8, which operate 0 50 100 150 200 250 300 350 400 450 500 550 600
MW
with similar exhaust temperatures to the FT60 IC, the last 126730.cdr
two stages will not need coatings, and maintenance action is Figure 1-4. Combined Cycles Cost Considerably
not needed for these stages at inspection intervals. At 4,000 More Than Simple Cycles
hr of full load operation and 400 starts per year, the first hot
1-2
gaspath inspection occurs at 6.25 years. The first major inspection occurs at 12.5 years, by which time, the airfoils
have accumulated 5,000 cycles and the HPT airfoils would need to be replaced.
Frame-type gas turbines are typically not designed for high-cyclic duty. In cyclic duty, E-class frame types can
typically last 3 years between hot gaspath inspections, where most turbine airfoils must be repaired and some
replaced. F-class frame types typically cannot go more then 2 years at 400 starts per year before a hot gaspath
inspection.
The cost to refurbish or replace a turbine blade is strongly correlated to the size of the blade, for comparable
materials and technology. The 1st-stage turbine blades of the FT60 IC are nearly the same size as those of a large
aircraft engine and much smaller then those of a large frame-type gas turbine. As an illustration, the sizes of the air-
foil sections of 1st-stage turbine blades are compared for the FT60 IC, the PW4090 large aircraft engine and an F-
class frame-type gas turbine in the same output class as the FT60 IC:
FT60 IC PW4090 F-Class Frame Type

AirfoiL Height (in.) 3.0 2.90 6.2
Axial Chord (in.) 1.10 1.15 4.20
We believe that O&M costs on a dollars/MW-hr basis will be favorable compared to those of frame-type gas
turbines.
1.3.3 Economic Analysis
As part of the market study, ADL did an economic analy-
Total Cost — Dollars/KW - Year
300
sis of the FT60 IC compared to the following frame-type gas
turbines: 7EA, 7FA, and a 2 × 1 combined cycle 7FX-2 (two 250
gas turbines and one steam turbine), representing a more 200

advanced combined cycle than the 2 × 1 combined cycle 7FA, 150
GT 7EA
CC: 7FX-2
NGT
which is widely used today. The performance of this power- GT FA
100 Coventional PC
plant is estimated at 590 MW and 58.5 percent net efficiency.
This was done to make a comparison against the types of 50
0 1000 2000 3000 4000 5000 6000 7000 8000
equipment that may be offered for peaking and intermediate Annual Operating Hours
126731.cdr
service at the time that the FT60 IC could be made available to
end users. A conventional pulverized coal powerplant was also Figure 1-5. NGT Is Competitive Over a Wide
used in the comparison. Figure 1-5 and Figure 1-6 show the Range of Operating Hours
results.
In Figure 1-5, the FT60 IC (NGT in the figures) provides the lowest cost of electricity (dollars/kW/yr) when
operated from about 500 to over 6,000 hr/yr. The peak savings occurs at 2,200 hr/yr (e.g. 8.8 hr/day, 5 days/week,
50 weeks/year), at which point the FA simple cycle and FX-2 combined cycle each cost $150/kW-year, while the
FT60 IC costs $130/kW-year. This amounts to costs of electricity of 6.8 and 5.9 cents per kW/hr respectively: a
savings of 13.2 percent. The savings are greater when compared to 1999 state-of-the-art instead of the anticipated
advanced combined cycle.
It is important to note that the combined cycle in this comparison has more than three times the output of the
FT60 IC, so the combined cycle enjoys a head start in economy of scale. The FT60 IC would have more flexibility
than the combined cycle in terms of the size of the commitment necessary on the part of the end user.
1-3
NGTs vs. FA GT in Peaking
NGTs vs. FA GT in Intermediate

Load
NGTs vs. FX-2 CC in Intermediate

Load
NGTs vs. FX-2 in Base Load
-5% 0% 5% 10% 15% 20%

Relative Economic Saving — percent
126732.cdr
Figure 1-6. Comparison of the Relative Attractiveness of NGTs By Calculating the NPV of
Production Cost/kW for Each Unit Over a 15-Year Investment Horizon
As an illustration of value to the end user, Figure 1-6 7000
shows that the FT60 IC provides savings of 8 to 18 percent
Annual Operating Hours

6500
compared to conventional gas turbines in net present value
(NPV) when operated in peak to intermediate load. 6000 FX-2 Zone
Figure 1-7 shows that the FT60 IC remains attractive in the
intermediate load range against the large combined cycle, 5500
even at high gas prices. 5000
NGT is competitive over a wide range of operating
4500 NGT Zone
hours for peaking and intermediate load applications; they
become even more attractive when operational flexibility 4000
(one key customer feedback) is factored in. 3.5 4 4.5 5 5.5
Fuel Price — Dollars/MMBTU
We compared the relative attractiveness of NGTs by 126733.cdr
calculating the NPV of production cost per kW for each Figure 1-7. Relative Competitiveness Between
unit over a 15-year investment horizon. NGT and FX2
1.4 RELIABILITY, AVAILABILITY, AND MAINTAINABILITY
The FT60 IC will be designed to accommodate cyclic 100
duty from the outset, which will be an advantage in reliabil- 98
ity compared to typical frame-type gas turbines in cyclic 96
duty. In addition to the turbine airfoil design discussed in 94
Section 1.3.2, the cases and disks will be designed for both 92
fast start and high-cyclic duty. 90
88 Frame Types
Another favorable attribute for the FT60 IC is that the
86 Aeroderivatives
core (HPC, combustor, HPT) section of the machine is
84
small enough to be readily shipped to a service shop for
82
maintenance while a replacement core is inserted to allow
80
the end user to continue operating. The core of the FT60 IC Reliability Availability
126734.cdr
is no larger that the FT8 gas generator, which is shipped
Figure 1-8. Aeroderivatives With Lease Engines
routinely to service shops in the same way. The FT60 IC
Have High Reliability and Availability
core is about 9-ft diameter maximum (at the DLE combus-
tor) and 8.4-ft long, while the GG8 (gas generator for a 25 MW aeroderivative) is 5.3-ft diameter and 13.3-ft long.
The core is similar to the gas generator of an aeroderivative in being the largest source of down time in the gas tur-
bine. This method of minimizing downtime is a proven way of improved availability for aeroderivatives, as illus-
trated in Figure 1-8.
1-4
The replacement core feature also improves maintainability, since the core can be shipped readily to a service
shop instead of requiring maintenance in the field. This is unusual for a gas turbine in the over 100 MW-size class.
1-5
2. RESULTS AND DISCUSSION
2.1 COMPONENT SYSTEM DESCRIPTIONS

This section of the report is a series of component system descriptions that are the result of test and evaluation
under the NGT Program, and were prepared by the design groups in PWPS who are responsible for those compo-
nents.
2.1.1 Compression System
The FT60 IC compression system shown in Figure 2-1 will consist of a 4-stage LPC and a 13-stage HPC,
employing the following state-of-the-art aerodynamic features.
• Low loss controlled diffusion airfoils
• Multiple circular arc (MCA) high Mach number rotors
• Radial bowed stators
• Contoured blade platforms
• Rotor tip stability bleeds.
The controlled diffusion airfoils will be designed with calibrated computational fluid dynamics (CFD) analysis
tools. The rotor system will incorporate an aerodynamic engine style rotor system to minimize rotor inertia to
achieve quick start to sync idle and full power. Low-cost material and construction techniques will also be incorpo-
rated to achieve low cost.
LPC 4 Stage
Intercooler Ducts
HPC 13 Stage
• No new technology in LPC/HPC for intercool cycle

• Material selection and configuration based on achieving low cost
• Mid compression intercooler allows lower cost HPC materials
• LPC and HPC designed to minimize hardware
• Rotors use aerodynamic-type construction to minimize inertia.
126735.cdr
Figure 2-1. FT60 Compression System
2.1.1.1 Low-Pressure Compressor

The LPC rotor construction (Figure 2-2) is planned to have a bolted 1st-stage disk/hub using an axial blade
attachment. The following three stages will be a welded drum with tangential blade attachments. The low 3,600
rpm N1 rotor speed allows use of a steel low rotor drum construction. This construction provides a low cost rotor
system that also minimizes blade attachment leakage and rotor inertia. Rotor rim cavities will be minimized and the
rotor shell placed as close as practicable to the inner flowpath wall to reduce the bending stresses on the tangential
blade attachment rails. The low compressor will have three stages of variable vanes with inner shrouds to reduce
vane deflection and bending loads on the variable vane trunnions. The slope or convergence of the inner flowpath
wall will be determined by both aerodynamic and cost concerns but will be set to achieve acceptable flowpath wall
2-1
All cases
Commercial available steel
PWA110 coating for
added corrosion protection
(36) (39) (46) (49)

(56) (58) (66) (69)
Rotor
AMS 6415 or PWA 768 steel
PWA110 or PWA77
Corrosion Coating
Blades and Vanes: Low Cost Rotor:

• Variable and fixed vanes AMS 5613 • Welded drum rotor (AMS 6415 or PWA768) steel
• Stator ID seal rings - AMS 6304 – Eliminates disk/spacer flanges
• PWA77 or PWA110 coating for – Eliminates integral disk wings or separate rim spacers
corrosion protection • 3 stages - Tangential blade attachments
• Blades AMS 5613 coated with PWA77 – Reduces leakage
for corrosion protection. – Reduces machining cost
126738.cdr
Figure 2-2. Low-Pressure Compressor Module
loading. The variable vanes will use large diameter outer circular platforms and large diameter vane trunnion posts
to minimize deflection and wear concerns at the variable vane bushings. The LPC case will be split axially at the
horizontal centerline for assembly purposes. This construction also serves to facilitate compressor field mainte-
nance. Materials selection is based on achieving low cost and meeting corrosion protection requirements. Disks are
planned to be wrought alloy steel PWA 768 or AMS 6415, and will require a corrosion protective coating of either
PWA77 or PWA110-2. The blades are proposed to be AMS5613 or 410 SST with the same corrosion coating used
on the disks. The compressor case will be cast AMS5613, 410 SST or 347 SST. Cases, if 410 SST, may also be
coated but this would be only for cosmetic purposes. The 300 series SST would maintain its appearance without
special coatings. This case will be heavy wall and not subject to stress corrosion concerns.
2.1.1.2 Intercooler Ducts
The intercooler ducts are planned to be cast quarter-segment ducts welded to form a transition from a full annu-
lus with a duct diverting flow to 180-degree horizontal ducts leading to the intercoolers. The exit and return ducts
have low-pressure loss aerodynamic and structural struts tying the inner and outer walls together and providing a
load path for the No. 2 and No. 3 bearing supports for the LPC rear bearing and the HPC front bearings, respec-
tively. Choice of material for the ducting remains under review. Candidate materials are cast aluminum and cast
410 SST. Cost and stiffness of these ducts will dictate material choice.
2.1.1.3 High-Pressure Compressor
The FT60 IC HPC (Figure 2-3) will be a scaled constant meanline compressor based on test experience with
our JT10D-2 and ATE compressors. The rotor will be a low inertia aerodynamic engine-style compressor to meet
quick start and low-cost objectives. The rotor aft end will be a welded drum with tangential blade attachments for
low cost. The drum construction reduces bolted flange interfaces and eliminates bolt hole machining and close tol-
erance snap fits. The elimination of these features also reduces customer maintenance cost for wear at these inter-
face locations. The shell of the drum will be located at the inner flowpath wall, thereby eliminating rotor rim
cavities. This construction requires that the fixed vanes be cantilevered, without inner vane shrouds. The cantile-
vered vane construction again offers opportunity to reduce stator costs. The forward fixed vane stages will require
inner vane shrouds because of the sloped inner flowpath forward wall. The forward drum will be a welded con-
struction similar to the aft drum. Tangential blade attachments will be used as far forward as practicable to reduce
machining costs. The rotor rim cavities, at the forward stages of the HPC, however, will limit use of tangential
blade attachments to the smaller blades in the aft end of the front drum. The rim cavities shift the drum shell inward
2-2
Cantilevered Fixed Stators

Split HPC Case AMS 5613 forward stages
Blades - Ams 5616 Cast PWA 1091 AMS 5616 rear stages
Greek Ascoloy
Bolted Rear Hub

Inconel 718
Drum Rotors
Steel PWA 768 Front drum - 5 Stages
Integral Hub R1 Disk
Steel PWA 768 Rear drum - 6 Stages
AMS 6415 or PWA 768 - Steel
Tinidur A286 Rear disk
• 19/1 pressure ratio compressor within P&W experience
• HPC will incorporate proven aerodynamic technology
• HPC designed with latest CFD tools
• HPC materials and construction to minimize cost (meet low dollars per kiloWatt goal)
• Case(s) axially split and double wall to provide:
– Easy maintenance
– Match blade tip response to meet rapid start time.
126736.cdr
Figure 2-3. High-Pressure Compressor
and thus increase drum rail rolling deflection, blade tip deflection, and results in thicker rails to reduce bending
stresses in the drum attachment rails. Further trade study of rotor inertia and drum weight versus cost is necessary
to determine the stage location where the switch over to axial blade attachments should occur.
The drum rotor construction requires the HPC cases to be split axially at the horizontal centerline for assembly
purposes. This construction also permits field maintenance of rotor and stator hardware, similar to the LPC plan.
An additional low cost and improved maintenance feature is planned for the fixed vane attachment in the HPC
case. The case will have 360-degree turned grooves to accept segmented tee-shaped vane retention hooks, which
will be radial and bolted to the case. These segments will be replaceable or serviceable for wear control. This will
eliminate wear refurbishment of the large split compressor cases and reduce customer maintenance cost.
The aft end stages, 7 through 13, will have a separate blade outer airseal case that will be structurally and ther-
mally isolated from the outer structural case. This double-wall case construction will allow the blade outer airseals
and case to respond to the flowpath temperatures to achieve the necessary blade tip clearance control required for
improved performance and quick start capability. The outer structural case will serve to carry the case bending
loads. The case stiffness will be matched to the forward intercooler, LPC case, and the aft combustor case. Rotor
and case dynamics will set stiffness and mass of these cases.
The intercooled cycle permits steel- or iron-based alloys to be used throughout the HPC. The plan, as with the
LPC, is to use AMS 5613 and AMS 5616 for blade and vane material. The rotor disks and drum is to be wrought
alloy steel, PWA 768 or AMS 6415. These materials will need corrosion protective coatings such as PWA 77 or
PWA 110-2 for all disks and blades. The compressor cases will be cast 410 SST and will also need corrosion pro-
tection for cosmetic purposes.
2.1.2 Combustion System
A robust, reliable, DLE combustor is key to the development of these NGTs. Technical issues associated with
DLE combustion chambers include combustion-induced pressure oscillations, limited cooling air, achieving near
perfect premixing of fuel and air, and maintaining nearly constant combustion reaction temperature over a wide
range of power settings and ambient temperatures. The NGT Program further exacerbates these issues by effec-
2-3
tively reducing the state-of-the-art in aerodynamic derivative NOx and CO emissions by a factor of two with
increased turndown capability, and improved start-up and response time.
The activities for the combustion system performed under Phase 1 of the NGT Program included the following:
• Conceptual design of the combustion system
• Cycle optimization for improved combustor turndown capability
• Definition of a proposed plan to mature the combustion system.
2.1.2.1 Combustion System Conceptual Design
The thermodynamic cycle selected for the NGT Program
Nox — 15% O2, ppmdv

100
features an overall pressure ratio of 5.57 Mpa (55 atm) and a
combustor exit temperature of 1510ºC (2750ºF) at standard
day base load operation. The NOx emissions characteristic 10
versus flame temperature of a typical premixer is illustrated

in Figure 2-4. The cycle, along with the program goals for 1
1400 1450 1500 1550 1600 1650 1700 1750
NOx emissions, dictates that approximately 95 percent of the Flame Temperature — °C
combustor airflow be used for premixing with the natural gas 126737.cdr
fuel prior to combustion.

Figure 2-4. Typical NOx Emissions Versus
The combustion system selected for the NGT Program is Flame Temperature
illustrated in Figure 2-5. The configuration is a fuel-staged
can annular combustion system. The configuration features a
primary combustion zone that serves to provide flame stabilization, light off, and low-power emission control. The
secondary combustion zone, defined by the position of the secondary fuel-air injection point, serves to provide a
distributed heat release for reduced combustion pressure fluctuations and a modulating flame temperature capabil-
ity for emissions turndown.
Figure 2-5. Conceptual Combustor Design
2-4
During typical operation, the primary combustion zone 105%

would be maintained at a nearly constant flame temperature, 100% Schedule 11
Normalized Combustor
Schedule 2
while the secondary combustion zone temperature would 95%
Exit Temperature
90%
vary over the gas turbine load range. The actual turndown
85%
capability of the combustion system is dictated by the com- 80%
bustor exit temperature lapse rate at part load conditions. A 75%
trade study has been conducted to evaluate the impact of 70%
combustor exit temperature schedule. 65%
60%
0% 20% 40% 60% 80% 100%
The compressor inlet guide vane position and LPC bleed Percent Load
fraction were the two main parameters selected for variation. 126740.cdr
Figure 2-6 illustrates the lapse of normalized combustor exit Figure 2-6. Effect of Compressor Inlet Guide
temperature, exit temperature normalized by the exit temper- Vane Modulation and Low-Compressor Bleed On
ature at 100 percent load, versus load. Combustor Exit Temperature Versus Load
Schedule 2 depicts the combustor exit temperature con- 1800

Schedule 11 - Primary Temp (°C)
sistent with 0-deg inlet guide vane position and minimum 1700
Schedule 11 - Secondary Temp (°C)
Schedule 2 - Primary Temp (°C)
LPC bleed. The resulting variation in combustor exit temper- Schedule 2 - Secondary Temp (°C)
1600
Flame Temperature — °C
ature is 287ºC (550ºF) between 40 percent and 100 percent
load conditions. 1500
1400
Schedule 11 illustrates the combustor exit temperature
consistent with a combination of inlet guide vane modulation 1300
and LPC bleed. The resultant combustor exit temperature 1200

variation is approximately 149ºC (300ºF). 1100
The combustion chamber flame temperature distributions 1000

0% 20% 40% 60% 80% 100%
for the two exit temperature schedules are presented in Percent Load
Figure 2-7. It is interesting to note that the same combination 126741.cdr
of flame temperatures can be achieved at 25 percent lower Figure 2-7. Flame Temperature Distribution
load with inlet guide vane modulation than without. Versus Load
2.1.2.2 Combustor Cooling and Material Selection
With 95 percent of the airflow dedicated to combustion, insufficient airflow remains to actively cool the com-
bustor structure. Therefore, a regenerative cooling system that first uses the combustion air to cool the structure
prior to mixing with fuel has been incorporated into the design. A detailed study of the regenerative cooling system
and selection of combustor materials is presented in advanced materials section of the report.
2.1.2.3 Proposed Plan to Mature Combustion System
The cycle operating conditions of the NGT Program present a significant challenge to the development of the
combustion system. Specifically, very few combustion facilities have the ability to simulate full engine pressure
conditions. The facilities possessing full pressure capability are extremely expensive and thus prohibit conducting
all of the development work at these conditions. Therefore, the proposed plan relies on conducting the majority of
development activities at reduced scale pressure, typically 2.03 Mpa (20 atm), with selected touch-the-wall confir-
mation tests conducted at full conditions.
2.1.2.4 Technical Approach
The technical approach proposed consists of focusing on three major challenges to achieving the program
emissions and reliability goals. The challenges are fuel-air mixture uniformity, thermoacoustic instability induced
pressure fluctuations, and combustion system durability. An approximate schedule is presented in Figure 2-8.
2-5
Activity
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Diffuser Airflow Uniformity
Fuel-Air Mixing
Mod Combustor Development
Design and Fabrication
Passive Dynamics Control
Active Dynamics Control
Full Combustor Development
Design/Fabrication Rig
Combustor Optimization
126742.cdr
Figure 2-8. Schedule for Development Activities

2.1.2.5 Diffuser and Combustor External Airflow Uniformity
A single can sector of the gas turbine combustion section will be fabricated using low cost stereolithography
models and Plexiglas. Atmospheric testing of the combustor diffuser section will be based on Mach number simu-
lation. Assessment of combustor section pressure distribution and airflow uniformity will be made through static
and total pressure measurements and flow visualization techniques. Heat transfer characteristics of the regenerative
cooling configuration will be validated by the use of single change temperature sensitive paint and semi-infinite
wall analytical techniques.
2.1.2.6 Fuel-Air Mixing

Spatial and temporal fuel-air mixing development will be conducted with planar laser-induced fluorescence
imaging using low cost stereolithography models of premixer designs. Steady and unsteady CFD analyses will be
employed to guide fuel-air mixing development.
2.1.2.7 Moderate Pressure Combustion System Development

A single can sector rig of the gas turbine combustion section will be designed and fabricated for testing at
UTRC. Combustion testing under moderate pressure conditions, 2.03 Mpa (20 atm), will be conducted to develop
combustion system performance. Assessment of combustor system performance will be made through measure-
ments of emissions, dynamic pressure, vibratory stress, and temperature distribution over simulated operating
range. Verification of premixer robustness to flashback/auto-ignition will be conducted through doping natural gas
fuel with higher order hydrocarbons.
2.1.2.8 Passive Combustion Dynamics Control

Convective Phase Angle Optimization
One and three dimension acoustic models of the combustion and fuel system will be constructed. Linearized
analysis of combustion system stability will be performed. Fuel injection location(s) will be defined. Combustion
rig tests will be conducted to verify predictions and guide selection of final fuel injection locations.
2-6
Acoustic Damper
One dimension model of coupled combustor, Helmholtz resonator system, will be constructed. Parametric
assessment of resonator designs will be conducted. Combustion rig tests will be conducted to verify predictions
and guide selection of final resonator design.
2.1.2.9 Active Combustion Dynamics Control

Actuator Sizing and Development
Valve bodies for use with MOOG drivers will be designed and fabricated. A set of MOOG model B74898 lin-
ear force motors and drive electronics will be procured. A test to determine dynamics response of actuator sized for
full mean flow will be conducted. Time-phased PLIF imaging of fuel-air mixture at the premixer exit plane in a sin-
gle-nozzle test rig will be conducted to quantify resulting fuel-air ratio modulation.
Fuel System Dynamic Assessment

An analytical model of the rig and engine fuel system to investigate fuel system response to the ACC system
will be developed. A mock-up of the rig and engine fuel system complete with actuation valves to verify analytical
predictions of fuel system response, will be constructed, if needed, based on analytical study results
2.1.2.9.1 Combustion Rig Demonstration

ACC components on the single sector combustion rig will be installed. Testing will be conducted to determine
the ability to control each of the acoustic modes individually and simultaneously.
2.1.2.9.2 Control System Integration

Design and construct a control system that is compatible with the gas turbine control system to ensure safe and
effective operation. Conduct simulations of the engine control system and the ACC system to analytically predict
system level response of the integrated system
2.1.2.10 Full Pressure Combustion System Development

A single can sector rig of the gas turbine combustion section will be designed and fabricated for testing at
Arnold Engineering Development Center. Combustion testing under full-pressure conditions will be conducted to
verify combustion system performance. Assessment of combustor system performance will be made through mea-
surements of emissions, dynamic pressure, vibratory stress, and temperature distribution over simulated operating
range.
The architecture of the combustion system shown in Figure 2-5 is to have separate combustion cans, each with
a separate convectively-cooled transition piece. The combustor control system uses compressor inlet guide vanes
to modulate the airflow to the combustor to achieve part load operation (Figure 2-6).
2.1.2.11 Physics Based Modeling of Combustors for Emission Predictions

One of the principal technical issues for meeting the NGT Program requirements is the development of a low-
emissions combustor. The high cost of combustor testing has lead to the incorporation of computer simulations into
the design process at P&W; design decisions are made from CFD predictions of flow splits, temperature exit pro-
files, and emissions. Therefore, the accuracy and reliability of the computational models are of crucial importance.
The coupled chemical and fluid-dynamic processes occurring within a gas turbine combustor are extremely
complex. Due to a huge range of both length and time scales within the turbulent flow field, direct numerical simu-
lations of the known governing equations are not feasible and will remain so into the foreseeable future. Thus,
physics-based models of these combustion processes are required.
2-7
A hierarchy of turbulence and combustion models exists based on model fidelity and computational efficiency.
To achieve reliable and accurate predictions of emissions in a gas turbine combustor over a variety of operating
conditions, there are three key requirements for the computational model:
1. High-fidelity representation of the combustor geometry (grid)
2. Turbulence model for highly turbulent, swirling flows with mixing
3. Combustion model that includes a chemical kinetics mechanism and a treatment of turbulence-chemistry
interactions.
The standard design process at P&W incorporates the use of three-dimensional, unstructured tetrahedral
meshes in its CFD calculations (Malecki, et al., 2001). With this grid topology, the complex geometric features
from diffuser exit to turbine inlet are readily modeled in a CAD-to-grid process that allows local grid refinement.
The closure of turbulence is achieved through standard approaches such as two-equation models (k-e, q-w), Rey-
nolds stress models, or large-eddy simulations (LES). It has been demonstrated that turbulence models at the Rey-
nolds-stress level provide improved results in strongly swirling flows over standard k-e models (Hogg and
Leschziner, 1989).
The calculated composition probability density function 30
(Composition-PDF) method is an advanced model for turbu-
lent combustion that incorporates arbitrarily complex chemical Inert
Reacting without kinetics
kinetics mechanisms without assumption in the highly nonlin- Reacting with ISAT
ear reaction term. Thus, it provides a natural treatment of tur- CPU Cost — Days 20
bulent-chemistry interactions. The solution of the
Composition-PDF, a high-dimensionality field, is calculated
through a statistical approach (Monte Carlo method) that uses
particles that are viewed as independent realizations of the 10
underlying turbulent flow. The particle code is coupled to stan-
dard CFD finite-volume codes through the heat release or den-
sity variables.
The computational costs of the Composition-PDF method 0
0.0E+00 1.0E+06 2.0E+06 3.0E+06
are higher than current moment closure combustion models, Cell Count
but recent hybrid finite-volume/Composition-PDF approaches 126743.cdr
(Muradoglu, et al., 1999; and Jenny, et al., 2000), coupled with Figure 2-9. Computational Cost Esimate for a
in-situ adaptive tabulation (ISAT, Pope, 1997) and automated Hybrid Finite Volume/Full Composition-PDF
reduced mechanism techniques, have significantly reduced Method
computational times. An initial estimate of the cost per unit as a function of cell count for such an approach is pro-
vided in Figure 2-9, which assumes a 16 species to 41 reaction-kinetics mechanism and parallel implementation on
48 Sun Sparc Ultra2s with 70 percent parallel efficiency.
Algorithm Development
The computational cost of Compositional-PDF methods has limited its application in combustor design calcu-
lations. With recent Composition-PDF algorithm advancements and the increase in computational speed of com-
puters, the development of a Composition-PDF approach with three-dimensional, unstructured grid capabilities
would mark a significant advance with potential to directly impact the combustor design process. The first step in
this development is the creation of a particle Composition-PDF code for the mixing of passive scalars. In inert
flows, the feedback coupling mechanism between the particle code and the CFD code (heat release) does not exist,
and the focus may be placed on the development and testing of efficient numerical algorithms and accurate physi-
cal models. The areas to be addressed may be classified as general physical/numerical issues concerning Composi-
tion-PDF methods, such as particle evolution, local time-stepping, correction algorithms, and mean flow
2-8
estimation; or as specific algorithms concerning three-dimensional unstructured grid implementation of the Com-
position-PDF method, such as particle tracking, domain decomposition, and particle boundary conditions.
General Composition-PDF Algorithms
The governing equation for the Composition-PDF varies in a high-dimensional space, which makes traditional
approaches for solving partial differential equations (e.g., finite differences) impractical. A Monte Carlo, or statisti-
cal, approach whose cost varies linearly with the dimensionality is required. The approach is based on stochastic
differential equations that govern the evolution of notional particles from which a Composition-PDF can be
derived. This density function is the modeled representation of the density function derived from the governing
equation.
For the Composition-PDF approach, the fundamental particle properties are the mass m*, position X*, and spe-
cies φ*, of the particle that evolve through
dm* = 0
1
§ · § 2 DT ·
2
* + 1 ∇D
dX ( t ) = ¨ U ¸¸ dt + ¨¨ ¸¸ dW
¨ ρ © ρ
T
© ¹ ¹
and dφ* = M(φ*,...)dt + S(φ*)dt

where Ũ is the Favre means velocity, ¢ ρ² is the mean density, and DT is the turbulent diffusion coefficient (Pope,
1985). These quantities are specified by the finite-volume code. Also, dW is an increment of an isotropic, vector-
valued Wiener process, and M and S are symbolic representations of the mixing and reactive processes, respec-
tively. In the current work, the mixing model implemented is the interaction by exchange with the mean (IEM)
model, while the reaction term is zero (non-reacting flow).
A two-step, mid-point numerical formulation was selected to provide second-order accuracy in the weak sense
for the particle position evolution equation. The particle species equation is solved by a first-order, splitting algo-
rithm. In this algorithm, the processes of mixing and reaction are split and calculated independently. To achieve
higher accuracy would require multiple evaluations of the reaction step, a computationally prohibitive prospect.
The complete algorithm is provided below.
Additionally, a local time-stepping algorithm for statistically stationary flows was developed and tested. In a
typical combustor grid, the grid size may range over an order of magnitude with relatively large skewness and high
expansion factors. The ability to scale the particle time step to the local conditions is critical to achieving an effi-
cient algorithm. The algorithm developed is detailed in Muradoglu and Pope (2001), and is based on constructing
an appropriate evolution equation for the normalized particle mass m* = m0µ*,
dµ* = βµ*dt
where β is defined with respect to a global time scale s such that
∂
ª µ ( s ) γ ( X [ s ] , s ) º¼ = 0
∂s ¬
where γ (x, s) ≡ ∆t (x, s)/∆s is the mapping from the global uniform time step to the local non-uniform time step.
The local time step is proportional to the minimum of the convective, turbulent diffusive, and turbulent time scales,
as defined from finite-volume CFD quantities. At steady state, the variable time-step (mass) modeled Composition-
PDF converges to the constant time-step (mass) modeled composition PDF; the local time-step formulation con-
verges in pseudo-time to the proper steady-state solution. The method was tested on a non-reacting bluff-body
flow, and was found by Muradoglu and Pope (2001) to “accelerate the global convergence of the hybrid method by
as much as an order of magnitude” and to “improve significantly the robustness of the hybrid method.”
2-9
As described in Muradoglu, et al. (2001), the hybrid finite-volume/particle PDF approach is consistent at the
level of the governing equations (or physical model), but may entail inconsistencies at the level of the numerical
algorithm. The inconsistencies arise from duplicate information contained within the particle and finite-volume
mean fields and can lead to unacceptable levels of numerical errors, particularly bias. Correction algorithms are
developed to force consistency. For the inert flows considered in a Composition-PDF approach, only the position
correction algorithm of Muradoglu, et al. (2001) is required. This algorithm enforces the consistency between the
time-averaged mean particle and particle mass densities (the particle mass density is defined by the volume of the
cell and the mass of the particles contained in it). The algorithm calculates a correction velocity Uc from a damped
oscillator equation driven by the normalized density difference and a correction potential calculated from the nor-
malized density difference.
The final discretized algorithm for the particle evolution is then expressed, with X0 ≡ X* (s), φ0 ≡ φ* (s), and
µ0 ≡ µ* as
1. Half step in position
1
° ª º½ ½° 2
X ≡ X + ®( 12 ∆t ) « U
1
2 0 + U C ( s ) + 1 ∇D » °¾ + °®( 1 ∆t ) 2 DT ¾ ξ
ρ ρ
T 2
°¯ ¬« ¼» °¿ X0 ¯° ¿° X0
2. Mixing of species
( ) 2
¬ ¼ ( ) ( )
φ M ≡ α X φ 0 + ª«1 − α X º» φ ∗ X
1 1
2
1
2
3. Reaction of species
(
φ * ( s + ∆s ) ≈ φ 1 ≡ R φ M , ª¬ ∆t , p º¼ X 1
2 )
4. Full step in position
1
° ª º½ ½
2
1 0 + U C ( s ) + 1 ∇D » °¾ + °® ∆t 2 DT °¾ ξ′
X ( s + ∆s ) ≈ X ≡ X + ® ∆t « U
ρ ρ °¿ X 12
T
°¯ ¬« ¼» °¿ X 12 ¯°
5. Full step in normalized mass

§ dt ( X1 ) ·
µ ( s + ∆s ) ≡ µ ¨
* 0
¸
¨ dt ( X 0 ) ¸
© ¹
Two independent Gaussian random vectors, ξ and ξ , with properties ¢ ξ² = 0 and ¢ ξξ² = I , are introduced.
The species-mixing step is presented in a general form to account for different mixing models. For the IEM mixing
model, the term α is
° ª C ∆t º ½°
( )
α X = exp ® « φ »
1
2
°¯ ¬ τ T ¼ X 1
2
¾
°¿
The reaction step for the species is represented by the reaction mapping R, which is a function of the mean
pressure and time step at the mid-step particle position and the mixed composition φM. The reaction mapping is
here zero, but would be provided from the implementation of chemistry through in situ adaptive tabulation (Pope,
1997).
2-10
The discretized particle algorithm requires properties interpolated to the particle position from mean values
specified by the finite-volume code or time-averaged from the particle properties. The algorithm that calculates the
particle means and the interpolation to the particles is based on trilinear interpolation, an approach well suited to
tetrahedral grids. Particle basis functions L provide the “weighting” for interpolation
4 3
q ( X* ) = ¦ Li ( X* ) Qi = Q4 + ¦ Li ( X* ) ( Qi − Q4 )
i =1 i =1
and for mean estimation
¦ µ * Ln ( X* ) Q *
Q n =
X ∈ Ωn
*
¦ µ * Ln ( X* )
X ∈ Ωn
*
where Ωn is the support for vertex n. The basis functions are calculated after each position evolution step from Li
(X*) ≡ Aij (X*j - xj(4)), where xj(4) is the position of the fourth vertex in the tetrahedron and A is the basis function
coefficient matrix that is constant within each cell.
Three-Dimensional, Unstructured Grid Algorithms
An efficient and robust algorithm is required to track the large
number of particles as they evolve through the grid; the particles must
be located within cells. The algorithm implemented calculates for each b X1
cell face the ratio of the face plane perpendicular distances from the
initial particle location and the face to the initial and new particle loca-
a
tions. If the minimum non-negative value exceeds one, the particle
remains in the cell. Otherwise, the particle exits the cell through the d2 / d1 > 1
d2
face with the minimum value for the ratio, and the process repeats (a
finite number of times) with the initial location moved to the cell face. X0 d1
An example in Figure 2-10 shows a particle leaving cell α and entering
cell β. Difficulties stemming from round-off error or high grid skew-
ness complicate the simple approach, but are handled robustly through 126744.cdr
error tolerances and identification of lost particles. The lost particles Figure 2-10. Particle Tracking Sketch
are discarded prior to further processing. In testing of a coarse grid
with a maximum number of face crossings set to 20, the probability of losing a particle in a given time step was
found to be less than 1e-7, a value deemed satisfactory.
Parallel processing through domain decomposition is required in finite-volume CFD codes to perform large-
scale combustor calculations. The ability to track a particle through a grid has to be extended to track a particle as it
moves from processor to processor. This algorithm inherently couples to particle tracking and particle boundary
conditions. Boundary and domain faces are tagged such that particles trying to cross these faces are specially
treated. Particles are reflected from inlet, wall, and symmetry boundaries, while particles are lost upon crossing exit
boundaries. Particles crossing periodic or domain boundary conditions are tagged as leaving the domain and posi-
tioned on the bounding face. The algorithm is designed so that these particles are bundled and broadcast to the
appropriate processor for continued tracking. This process is repeated a fixed number of times prior to labeling the
remaining unlocated particles as lost. The coding for the parallel processing and the interdomain face connectivity
is highly dependent on the data structures implemented in the finite-volume CFD code. The coding for parallel pro-
cessing was not performed under this contract, while the framework for the domain-crossing boundary conditions
was coded. Therefore, the current composition PDF code is limited to single domain calculations.
2-11
The final algorithm implemented is for the treatment of

particles entering through inlet boundary conditions. The algo-
rithm mirrors the two-step, mid-point algorithm by specifying
a fixed number of particles to have a mid-point position on the
bounding surface. These particles are assumed to have initi-
ated from a location perpendicular to the face plane and at a
distance uniformly distributed. These particles then evolve a
full particle step. Some of the particles will enter the domain
and are kept, while others are lost. The mass of the entering
particles is specified so that the inlet mass flow rate is held
constant, while the particle species are set to the inlet condi-
tions.
Figure 2-11. CFD Model and Grid
Jet-in-Crossflow Test Case
A jet-in-crossflow (JIC) test case was constructed to test
the implementation of the Composition-PDF code, but not nec-
essarily provide accurate solution to the physics. As such, the
grid is rather coarse with 4.5K cells and with a plane of sym-
metry through the jet center. The CFD model is contained in
Figure 2-11 with surface grids illustrated. The configuration
consists of a fuel jet into a crossflow of air with a momentum
ratio of 25. The case is nonreacting, and the species variable
may be viewed as a mixture fraction. The CFD mixture frac-
tion profiles are illustrated in Figure 2-12.
Four Composition-PDF cases were run with initially 20,
40, 80, and 160 particles per cell. The results from all four Figure 2-12. CFD Calculation Mixing Profile
cases were comparable and only those from the 20 particles
per cell case are shown here. Figure 2-13 plots, as a function of iteration, the instantaneous exiting particle mass
flow rate and its iteration average (range 2000 to 7200), both normalized by the imposed CFD mass flow rate. As
expected, the iteration average matches the imposed CFD mass flow rates. The instantaneous mass flow rate oscil-
lated with a standard deviation of 2 percent for the 20 particles per cell case. The standard deviation was found to
scale with 1/N, the inverse of the number of particles per cell, as expected in the Monte Carlo algorithm.
Figure 2-14 and Figure 2-15 contain the mean mixture fraction and mixture fraction variance at a vertex located at
the leading edge of the jet about one jet-diameter into the crossflow and slightly off-center (see the red circle in
Figure 2-11). Plotted are the instantaneous mean, the running time average over 500 iterations calculated within the
Figure 2-13. Convergence of Exit Mass Flow Rates Figure 2-14. Convergence of Mean Mixture
Fraction at Selected Vertex
2-12
Composition-PDF code, and the post-processed iteration aver-

age (range 3,000 to 7,200). Clearly, a statistically stationary
state is reached for this vertex located in an active region of the
flow, but a longer time-average scale should be used in the
Composition-PDF calculations. In addition, the computational
time per particle per time step that was found in these calcula-
tions was comparable to the values used in the estimates pro-
vided in Figure 2-9. The final results presented are the time-
average mean mixture fraction contours in Figure 2-16. These
results show a significant difference from the CFD calculations
in Figure 2-12. A possible explanation for the discrepancy is
that the Composition-PDF code is more dissipative than the
CFD code, which on such a coarse grid leads to unacceptable
Figure 2-15. Convergence of Mixture
errors. This issue must be investigated further.
Fraction Variance at Selected Vertex
2.1.3 Turbine System
During the NGT Program Phase I study, several concepts
were evaluated for the turbine. The major decision to make
concerned whether to have a single or two-stage turbine. After
detailed aerodynamic studies, and a realization of the common-
ality requirement between the NGT family of engines, a two-
stage turbine was selected. A common 50 Hz and 60 Hz power
turbine was also selected due to the cost savings from eliminat-
ing a gear box. The commonality would be achieved by design-
ing to a speed between 3,000 and 3,600 rpm while featuring
different sets of airfoils on both applications. A list of technolo-
gies has also been identified. The funding for these technolo-
gies comes from DOE, NASA, the U.S. Air Force, and P&W. Figure 2-16. Composition-PDF Mixing
2.1.3.1 Turbine Material Selection
Ceramic vane technology will require less cooling flow than conventional metallic vanes and will therefore
improve the engine efficiency.
Block No. Stage Material Comments
Block 1 1st-stage vane Cobalt
2nd-stage vane, 3rd-stage vane ds Nickel Ceramics not at TRL 6
1st-stage blade, 2nd-stage blade, 3rd-stage blade DS Nickel May also use advanced IGT alloy
4th-stage vane through 7th-stage blades EQ Nickel
Block 1.5 2nd-stage vane, 3rd-stage vane Si3N4 Uncooled 2nd-stage vane and 3rd-stage vane
Block 2 2nd-stage vane, 3rd-stage vane Si3N4 Uncooled
1st-stage vane, 1st-stage blade, 2nd-stage blade, DS Nickel May also use advanced IGT alloy, or cooled Si3N4
3rd-stage blade 1st- stage vane
4th-stage vane through 7th-stage blade EQ Nickel
The HPT disk material will likely be IN 718, although low-cost material such as M152 steel will also be evalu-
ated. LPT disk material will most likely be steel, with Nickel-based alloy as the back up.
2-13
2.1.3.2 Turbine Overhaul and Maintenance

No coating on later stages of LPT provides cost savings during overhaul and maintenance.
Stage Coating Stage Coating
1st-stage vane Low-k TBC 1st-stage blade Low-k TBC
2nd-stage vane Low-k TBC 2nd-stage blade Low-k TBC
3rd-stage vane Low-k TBC 3rd-stage blade Low-k TBC
4th-stage vane Oxidation 4th-stage blade Oxidation
5th-stage vane Oxidation 5th-stage blade Oxidation
6th-stage vane None 6th-stage blade None
7th-stage vane None 7th-stage blade None
2.1.3.3 Phase II Turbine Plan

The turbine plan for Phase II of the NGT Program is as follows:
• Evaluate potential turbine technologies for incorporation into FT60IC
— Ceramic vanes, ceramic matrix composites (CMCs), microwave sensor ACC, trailing edge optimiza-
tion, endwall contouring, advanced IGT alloy
• Conduct detail analyses of FT60IC turbine for initial and final cycles
— Design (e.g. rotor sizing, airfoil designs, etc.)
— E&D and Engine Cost
— Engine Commonality (FT60IC, FTX, FT60)
• Develop engine designs that incorporate advanced technologies
— Conduct rig and demonstrator tests.
Conceptual Design Tasks: Preliminary Design Tasks: Detail Design & Manuf. Tasks: Validation Tasks:
- Preliminary Cross-sections - Flowpath Optimization - Final Airfoil Definition - Test Support
- Configuration Selection - Work Split Optimization - Airfoil Casting Definition
- Technology Evaluation - Cost Optimization - BOAS Design & Manuf.
- Technology Selection - Clearance Assessment - Blade Design & Manuf.
- Material Characterization - Risk Assessment - Vane Design & Manuf.
- Explore potential cost reductions - Preliminary Airfoil Definition - Disk & Accessories Design & Manuf.
- Validation Plan - Material Trade Studies and Selection - TOBI Design & Manuf.
- Disk Forging Sizing - Finished Cross-section
- Bill of Material
Conceptual Design Preliminary Design Detail Design & Manufacturing Validation

2002 2006 2007 2009
126751.cdr
Figure 2-17. Timeline and Tasks for Block 1 FT60IC
2-14
Figure 2-18. FT60IC Turbine Cross-Section
Table 2-1. NGT Program Turbine Technologies1

TRL 6 Funding Funding DOE Funding Outside
Technologies Date Program Source Required? Resource
Ceramic Vanes With EBC 2005 NGT DOE Y2
Microwave Sensor Active Clearance Control 2004 Internal Internal Y Universities
Film Optimization With CFD and Leading Edge Fillet/Endwall 2005 IHPTET U.S. Air Force, Y Universities
Contouring (PRDA VI) Internal
Trailing Edge Optimization 2003 Internal Internal Y
Tip Desensitization – Tip Blowing and Bowing 2005 IHPTET U.S. Air Force, Y Universities
(PRDA IV) Internal
Advanced IGT Alloy 2006 ?
CMC Blade Outer Airseals ?
Aggressive T-Duct 2005 UEET NASA N
Hi-Life Airfoils 2005 UEET NASA N Universities
Multidiscipline Optimization 2002 Internal Internal N
Low-Conductivity Thermal Barrier Coating 2005 Internal Internal N
1
DOE funding is required for all technologies, except for advanced IGT alloy and CMC BOAS whose requirements are currently
unknown.
2 Funded through 2001.
2-15
2.1.4 Ceramic Turbine Vanes and Ceramic Matrix

Composite Liners
Under Phase I of the NGT Program, UTRC supported
P&W by studying incorporation of ceramic turbine vanes
and CMC combustor liners/transition ducts in the FT60 IC
engine. A cross-section of the FT60 IC combustor and tur-
bine sections is presented in Figure 2-19. The results of this
initial study show that ceramic turbine vanes and CMC com-
bustor liners/transition ducts will help the FT60 IC meet the
25,000-hr/4,000-cycle life requirement. The performance
and cost benefits of using ceramics and CMCs over metallic
systems are discussed below.
2.1.4.1 Ceramic Turbine Vanes
During Phase I, ceramic turbine vanes were identified as
Figure 2-19. FT60 IC Combustor and Portion
an advanced material technologies that is key to the high
of the Turbine
efficiency and high performance of the FT60 IC. The Phase I
objective was to assess the feasibility, benefits, and risks of using ceramic 1st-, 2nd-, and 3rd-stage turbine vanes.
A material database was first established to summarize the thermal and mechanical performance of silicon nitride.
Secondly, a durability study was carried out to assess the cooling flow required by ceramic turbine vanes, as well as
their metallic counterparts, based on the material properties and the temperature envelope defined by the concep-
tual design of the turbine section. Following the durability study, a business case study identified the benefits asso-
ciated with the cooling flow saved as a result of using ceramic turbine vanes.
Material Database
Thermal and mechanical properties of the potential turbine vane materials are required in order to perform a
durability study. Therefore, the first part of the work was to collect thermal and mechanical data of commercially
available silicon nitride that was developed for high temperature structural applications. The candidate silicon
nitride materials were AS800 from Honeywell Ceramic Components and SN281 and SN282 from Kyocera Indus-
trial Ceramics Corporation. Thermal and mechanical properties were gathered by collecting data from vendors,
from literatures, and from research institutes (such as Oak Ridge National Laboratory, University of Dayton
Research Institute, and NASA Glenn Research Center) that have been studying these materials. With the existing
mechanical properties and the temperature envelope defined by the conceptual design of the turbine section, a pre-
liminary analysis was carried out to identify the stress envelope in which the material would meet the 25,000-hr life
goal. It was concluded that with a design specified for the ceramic material, it is possible to manage the thermal
stress in the turbine vane within the stress envelope (less than 30 ksi). Mechanical properties that are critical to
ceramic turbine vane design and life prediction were also identified. It was recognized that additional slow crack
growth and long-term creep rupture data was required. Such findings have been communicated to the teams at
ORNL, UDRI, and NASA that are currently conducting mechanical testing on structural ceramics.
Durability Study
According to the temperature envelope defined by the conceptual design of the turbine section and the material
properties, it became clear that the 2nd- and 3rd-stage ceramic turbine vanes in FT60 IC would not require cooling.
About 8 percent of the total turbine cooling flow could be saved by replacing the 2nd- and 3rd-stage metallic tur-
bine vanes with ceramic turbine vanes. The 1st-stage ceramic turbine vane still requires cooling due to the high-
operation temperature. A durability study was carried out to assess the cooling flow requirement for the 1st-stage
ceramic turbine vane versus its metallic counterpart. The analysis of the cooled 1st-stage ceramic turbine vane was
largely based on the knowledge and experience gained under previous programs that developed the cooled silicon
nitride FT8 turbine vane, while the analysis on the cooled 1st-stage metallic vane was based on existing P&W tur-
bine airfoil designs. It was found that a cooled 1st-stage ceramic turbine vane would require only convective cool-
ing, while a state-of-the-art metallic turbine vane requires film cooling. The cooling flow required by a cooled 1st-
2-16
stage ceramic turbine vane was only one-fourth to one-third 0.8

of what was needed by a metallic turbine vane, as shown in 0.7 1 Wceramic 1
PHI = (Tg - Tm)/(Tg - Tc/a)

á á
Figure 2-20. Using cooled 1st-stage ceramic turbine vanes 4 Wmetal 3
0.6
could save about 33 percent total turbine cooling flow. Over- Film
0.5
all, a turbine section with ceramic 1st-, 2nd- and 3rd-stage Cooled
0.4 NGT Metal IV
vanes will use only the equivalent of 60 percent of the cooling
airflow a turbine with state-of-the-art metallic vanes will use. 0.3 NGT Ceramic IV
A preliminary business case study has indicated that such 0.2
savings in turbine cooling flow could result in more than a 0.4 Convective
0.1
Cooling
percent improvement in overall thermal efficiency and 0
increase the power output by the equivalent of 7 percent. 0 1 2 3 4 5 6 7 8 9 10
Cooling Flow
126754.cdr
2.1.4.2 CMC Combustor Liner and Transition Duct
Figure 2-20. Cooling Flow
CMC combustor liners and transition ducts have been
evaluated for the FT60 IC combustor, as shown in
Figure 2-19. The CMC system being evaluated is melt-infiltrated silicon carbide fiber-reinforced silicon carbide
(MI SiC/SiC) because it is the most mature CMC system and has the temperature capability and conductivity
required in combustor environments. MI SiC/SiC has demonstrated long-term stability at temperatures up to
1204°C (2200°F), and efforts are underway to extend the use temperature of MI SiC/SiC to 1316°C (2400°F). In
order to inhibit recession of the SiC in the high-temperature, high-moisture combustor environment, an environ-
mental barrier coating (EBC) is applied to the hot surface of the SiC CMC combustor components. The current MI
SiC/SiC EBC system has a use temperature of 1316°C (2400°F), and work is being conducted to extend the use
temperature to 1482°C (2700°F).
Thermal Analysis
Heat transfer, surface temperature, and pressure drop calculations for the preliminary design of the FT60 IC
combustor liner were performed using the P&W P783 General Internal Flow Solver Program. The purpose of this
preliminary design study was to examine the tradeoff between the temperature capability of a number of potential
liner materials and the associated coolant-channel pressure drop required to maintain acceptable material tempera-
tures. In other words, liners with higher temperature capability will require less cooling which, in turn, implies
lower velocities and reduced pressure drop in the cooling channel. The key result is that the CMC liners had a 5
percent higher static pressure than the metal liners, which translates into a 1 percent pressure drop savings for the
CMC liner based upon P&W design analysis. This 1 percent pressure drop savings translates into a 0.5 MW
increase in power according to the P&W Engine Performance Group.
The maximum interface temperature for the MI SiC/SiC liner was approximately 982°C (1800°F). There is a
large margin of operating temperature safety for MI SiC/SiC at 982°C (1800°F), since this system has demon-
strated long-term durability at 1204°C (2200°F). The interface temperatures for the metal liners are 815°C
(1500°F) to 871°C (1600°F). These are close to the temperature limit for the various metals, once a 38°C (100°F)
safety margin is included for hot spots.
Durability Analysis
A preliminary durability analysis has been performed on candidate designs and materials for the combustor
liner and transition duct. Several materials were considered, including MI SiC/SiC, Nickel B-1900 (PWA1455),
and Nickel SC2000 (PWA1487). To improve life, the CMC and metal components featured EBC and TBC coat-
ings, respectively. The candidate design and materials were analyzed to determine:
• An operational stress-temperature envelop
• A relative comparison of inner/outer wall strains
• EBC durability.
2-17
The operational stress-temperature analysis produced a B-basis life curve that predicted that MI SiC/SiC can
operate continuously for 25,000 hours under combined temperatures and stress conditions of up to 1148°C
(2100°F) and 15 ksi. Calculation of the hoop stresses resulting from through-thickness thermal gradients indicates
that the CMC liner/EBC system will have approximately one-third the stress of metal liner/TBC systems. Analysis
of the operating stress for the CMC liner/EBC system and the various metal liner/TBC systems indicates that all
systems will achieve the 25,000-hr goal; however, the CMC liner/EBC has a larger margin of safety.
The relative strain due to thermal expansion differences between the combustor liner and the shell was approx-
imately half the value for a CMC liner, compared to a metal liner. The actual stress in the liner will be a function of
the attachment scheme; however, the lower relative strain has the potential for resulting in lower liner stress and
simplified attachment design.
The EBC coating should meet the 25,000-hr life goal for the FT60 IC, and EBC development work at UTRC is
expected to further reduce the recession rate. The recession rate of 0.012 microns per hour was measured on an
EBC-coated MI SiC/SiC liner in a Solar Turbines Centaur 50S engine run at the Texaco site in Bakersfield, Califor-
nia, from 1999 to 2000. Therefore, 300 microns of EBC would be recessed in 25,000-hr, which is less then the
thickness of the coating.
Business Case Study Additional Revenue
Additional Fabrication Cost
As discussed above, ceramic turbine vanes and CMC
combustor liners/transition ducts required less cooling 1
and increase engine efficiency. This increased efficiency
means that the same basic engine will produce more
Increased Revenue
power and be able to be sold at a higher price. However,

the ceramic components are projected to cost more then
and Cost
metallic components. Figure 2-1 shows the business case

study results.
The increased revenues shown in Figure 2-1 are
slightly offset by the increased fabrication cost on manu-
facturing ceramic and CMC components. Even with this
0
increased fabrication expense, a significant revenue 2nd- and 3rd-Stage 1st-, 2nd-, All Ceramic
Ceramic Vanes and 3rd-Stage Vanes and
increase is projected for the FT60 IC with all ceramic tur- Figure 2-1. Increased Revenue WithCombustor
Ceramic Vanes Ceramic and
Liner
bine vanes and CMC combustors/transition ducts. CMC Components 126861.cdr
2.1.5 Mechanical Systems

Initially, when the gas turbine was to be derived from
an aircraft engine, the bearings were to be a rolling ele-
ment. The high axial thrust load, caused by the high pressure level of the intercooled cycle, would have required
the development of advanced tapered roller bearings for the low-pressure rotor. When the size of the gas turbine
was increased, so that the airflow was in the same class as older (E-class) direct-drive frame-type gas turbines, the
bearing choice became conventional hydrodynamic bearings, with journal bearings for the radial load, and tilting
pad bearings for the axial load. This decision is favorable for long life (RAM) and reduces risk in the NGT Pro-
gram.
2.1.5.1 Balance of Plant Summary
PWPS is studying the development of an intercooled gas turbine derived from P&W's FTX engine program.
This PWPS development program has identified the land-based intercooled gas turbine generator set as the FT60
IC. Several intercooler configurations and corresponding piping alternatives were investigated in support of this
development program. These studies determined FT60 IC performance and simple-cycle equipment installed costs
for various intercooler configurations.
2-18
2.1.5.2 Lubrication
The FT60 lube oil system contains the engine lube oil system together with necessary filters, cooler, and reser-
voir.
Components not integral to the gas turbine are mounted as a pre-engineered package on the ancillary skid.
This package, located adjacent to the gas turbine enclosure, contains the combined reservoir, duplex filters, air-to-
oil cooler, and the turbine supply and scavenge pumps.
The turbine lube oil system lubricates and cools the main bearings and bearing seal assemblies. Lubrication is
provided by identical three-element pumps, which provide primary and auxiliary flow functions. The primary and
auxiliary pumps are alternating current motor driven at 3,600 rpm, while the emergency pump is a direct current
motor driven at 1,800 rpm. The auxiliary pump automatically starts in case of trouble with the primary pump. The
direct current pump is used for emergency shutdown when alternating current power is lost.
2.1.5.3 Gas Fuel System
The customer-furnished fuel gas supply is provided at a minimum pressure. After flowing through a fire valve,
the gas flows through two shutoff valves and the modulating valve to the gas manifold, where it is injected through
the gas fuel nozzles. The modulating valve meters fuel in response to signals from the electronic gas turbine con-
trol.
2.1.5.4 Liquid Fuel System
In the liquid system, fuel in a flooded suction line from the fuel storage tanks is pumped by the customer-sup-
plied fuel forwarding system to the filtering skid. After flowing through a fire stop valve, the fuel is boosted by fuel
pump, and sent to the modulating valve. The modulating valve meters fuel in response to a signal from the elec-
tronic gas turbine control. The fuel then flows through two fuel shutoff valves and a flow divider, into the manifold
and through the liquid fuel nozzles.
2.1.5.5 Dual Fuel System
The dual fuel system uses the same components as the gas and liquid fuel systems, plus additional valves to
allow switching from one fuel to the other.
At startup with liquid fuel selected, the fuel pump is started and liquid fuel flows through the liquid modulating
valve and liquid shutoff valve into the liquid fuel manifold to the nozzles. At shutdown, the pump is disengaged,
the liquid modulating valve and liquid shutoff valves are closed, the liquid drain valves are opened to prevent pres-
sure buildup downstream of the shutoff, and the manifold drain valve is opened to drain the manifold.
When gas fuel operation is selected, gas flows from the supply through two shutoff valves, through the modu-
lating valve to the gas manifold. A small amount is sent through the liquid system purge valve to the liquid fuel
manifold for continuous purging of all liquid fuel from the manifold and nozzles to prevent coking of the nozzles.
At shutdown, the shutoff valves are closed the vent line between shutoff valves is opened to relieve pressure.
The gas turbine has the capability to burn liquid and gaseous fuels simultaneously or to transfer from one fuel
to the other during operation.
2.1.5.6 Water Injection System
The water injection system (used for liquid fuel) introduces demineralized water with the fuel into the nozzles
of the gas turbine to assist in achieving required exhaust gas emission levels. The system operation is automatically
controlled by electronic gas turbine control.
The demineralized water is delivered by the customer to the water injection skid located just outside the gas
turbine enclosure. The skid contains the alternating current motor driven pumps to boost the water to the required
pressure and necessary valving.
When operating on liquid fuel, the water is mixed with the fuel and sent through the liquid fuel nozzles. When
running on gas fuel, the water is sent through the liquid fuel nozzles and the gas through the gas fuel nozzles. When
2-19
running on both fuels, the water is mixed with the liquid fuel and injected through the nozzles via the liquid fuel
manifold, while the gas flows through the as fuel nozzles.
2.1.5.7 Starting System
The starting system consists of an electric starter mounted on the gas turbine.
When the operator initiates a start, the power is supplied to the starter motor geared to the high-compressor
rotor shaft of the gas generator. The starter converts the fluid pressure to shaft torque and rotates the high-speed
rotor to ignition speed.
After fuel is admitted to the gas turbine combustion section and ignition is achieved, the gas turbine accelerates
to its self-sustaining speed and the starter is disengaged from the shaft. If any problems are detected during the start
cycle, the control shuts off fuel flow and hydraulic fluid flow, causing the starter to disengage and the gas turbine to
coast down.
The starting system is also used to rotate the gas turbine rotors for water wash/gaspath purge. When water wash
is selected, the start system is used to motor the gas generator while water, with or without detergent, is sprayed
into the bellmouth. During a purge operation, fumes and other flammable gases accumulated in the gas turbine are
purged by motoring the gas generator with the starting system to above 1,500 rpm, with the ignition and fuel sys-
tems off.
2.1.5.8 Fire Protection System
The fire protection system provides independent fire detection and CO2 total flooding fire suppression systems
for the gas turbine enclosure and is available as an option for the generator and control enclosures.
Automatic fire detection is provided by rate compensated thermal detectors. Facilities for manual (electric and
mechanical) initiation of the fire systems are also provided. The CO2 tanks, solenoids, and manifold are located
outside the enclosures, with the system's control module located in the control room.
The control system monitors and displays the status of all fire system inputs and provides outputs to activate
audible and visual alarms, discharge CO2, close fuel fire safety valves, and signal turbine and unit control systems
for required responses. The system operates on 24V of direct current and contains its own internal power supply
and battery backup.
Immediately upon actuation of the system, the turbine enclosure secondary air supply fans are de-energized
and the fuel supply is shut off. A 20-sec time delay permits rundown of the gas turbine and generator before a sole-
noid valve releases the pressurized CO2 into its distribution manifold. A pneumatic cylinder, actuated by the pres-
surized CO2, releases a pair of guillotine-type dampers (just forward of the fan blades), closing off the secondary
air path. Simultaneously, a series of nozzles floods the enclosures to a 34 percent CO2 concentration, sufficient for
inerting the combustion process.
The CO2 supply to the manifold is fed from pressurized tanks. The first tank is quick emptying, while the sec-
ond slow-emptying tank maintains the five percent level required to overcome dilution from air leakage. This CO2
concentration is maintained for approximately 30 minutes, sufficient time to allow combustibles to cool below
their autoignition temperatures.
A CO2 status display board is provided near each protected enclosure entry to visually indicate the status of the
fire protection system (CO2 armed or disarmed).
Disarming may be accomplished by disabling the CO2 discharge system either electrically by means of a key-
switch and/or blocking the flow of CO2 by a manually-activated safety block valve in the CO2 piping discharge
system. When disarmed, the detection alarm system will remain active while the CO2 discharge capability will be
disabled. Continuous alarm signals are sent to the monitoring system, notifying the operator while the system is
disarmed. Additional safety features include a suppressant release delay and audible and visual alarms inside and
outside the enclosure.
2-20
2.1.5.9 Gas Detection

Gas detection in the gas turbine enclosure is provided by a resistance-type combustible gas sensor. When the
gas concentration reaches 20 percent lower flammable level (LFL), the gas hazard alarm will be displayed in the
fire control panel and warning alarms will be activated at the enclosure. When a 60 percent LFL is reached, an
automatic trip of the fuel and gas turbine will be initiated. The enclosure’s ventilation system will remain inopera-
tion to reduce the gas hazard.
2.1.5.10 Gas Turbine Dehumidifier System
The gas turbine dehumidifier maintains the internal parts of the gas turbine at a temperature above the dew
point of the ambient air during nonoperation periods. Dry air forced into the compressor section and out through
the exhaust ducts prevents condensation, which can cause corrosion.
The operation of the humidifier system is fully automatic for unattended operation.
2.1.5.11 Water Wash System
The gas turbine requires periodic washing of its aerodynamic components that have accumulated deposits that
could affect performance.
A water wash circuit, when activated by the operator, energizes a solenoid valve in the water wash line. The
water flows through a nozzle located inside the inlet air plenum, directed toward the gas turbine's inlet. While the
gas turbine is rotated on the starter, the sprayed water is pulled through the compressor section of the gas turbine
and is drained through drain valves. After washing, the gas turbine is started to dry out any remaining water.
2.1.5.12 Electric Generator
The electric generator is an open ventilated, two-pole, air-cooled unit rated to BS/IEC standards with Class F
insulation, but limited to Class B temperature rises on a total temperature basis. It includes shaft-mounted overhung
main and pilot-brushless exciters complete with rotating-fused diodes, and all required support auxiliaries, instru-
mentation, protective devices, and controls.
The rotor is an integral forging of nickel chromium molybdenum alloy steel. Axial slots are milled on the
periphery of the body of the rotor to carry the winding and for ventilation. The rotor winding conductor material is
conventional silver bearing copper. The conductor is in the form of a strip and each rotor coil is preformed to the
shape required. All insulation materials are suitable for Class F operation.
The preformed coils are inserted into the slots, each turn being insulated from the next. After the completion of
the winding, the conductors are heated electrically and pressed to the correct depth using pressing rings. A fully
interconnected damper winding is then fitted into the tops of the slots and the retaining wedges are inserted. The
rotor endwinding is braced with packing blocks between the conductors, after which the nonmagnetic manganese
chromium steel endcaps are shrink fitted to spigots at each end of the rotor body. All rotors are tested at 20 percent
overspeed and balanced.
The main bearings are conventional circular profile, white metal lined, hydrodynamic cylindrical bearings.
Pressurized oil seals are fitted at each end of the bearings. Temperature detectors are provided for the bearing metal
and oil drains. Noncontact vibration sensors are provided for both bearings.
The generator features integral axial flow fans which supply a large volume of filtered cool air flowing through
the frame and over the stator and rotor coil ends and winds through interslot ventilation ducts and discharging at the
top of the frame. The exciter is self-ventilated.
A separate air-cooled lubrication system is provided, including a storage tank with radiator and cooling fan,
two alternating current motor driven lube oil pumps, and a direct current motor driven emergency backup pump for
emergency coastdown or black start.
The generators are fitted with a brushless excitation system, where the excitation power is derived from a small
alternating current generator driven by the generator shaft. The alternating current power produced by this exciter
2-21
is rectified to direct current by a shaft-mounted rotating rectifier assembly that is connected to the main field
through conductors inside the shaft; thus, the need for sliprings is eliminated.
2.1.5.13 Electrical/Control System
The electrical/control system contains all of the equipment necessary for local control of the FT8 60 IC
together with the switchgear and generator protectives.
The equipment contains the gas turbine and generator controls, motor control center, low voltage alternating
and direct current distribution, station batteries, metal-clad switchgear, station auxiliary transformer, protectives
relaying, and master terminal board. These controls and instruments are mounted on the following cabinets:
• Protective Relay Cabinet
• Instrument Cabinet
• Operator Cabinet
• Motor Control Center
• Monitoring Cabinet
• Master Terminal Board Cabinets.
2.1.6 Control System
The control system contains an integrated gas turbine and unit control. This integrated controller acts as a cen-
tral processing point for all input/output serial and ethernet communications associated with the FT60 IC. Data
from this control is sent to a user-friendly ICE operator interface to display pertinent information.
The Woodward controller performs both fuel control and sequencing functions. This system incorporates a dig-
ital programmable microprocessor that optimizes turbine safety and efficiency. The programmable features of this
control enhance the ability of the end user to incorporate the latest features in gas turbine technology.
This integrated control performs all of the gas turbine control functions including:
• Speed Control
• Temperature protection and control
• Acceleration and deceleration limiting
• Fuel valve control
• Inlet guide vane control
• Variable stator vane control
• Water injection
• Gas turbine performance monitoring
• Start sequencing
• Unit synchronization
• Alarm and shutdown protection
• Monitoring System.
The ICE monitoring system package operates on a Pentium-based industrial computer and serves as the opera-
tor interface. This system interfaces to the gas turbine and unit control through an Ethernet local area network. The
monitoring system performs such functions as:
• Data logging and trending
• Alarm monitoring
• Alarm and event logging
• Sequence-of-events recording
2-22
• First-out alarm indication

• X-Y plotting
• Calculation functions
• Event storage, archiving, and redisplay
• Operator control functions
• Process animation
• Control system diagnostics.
2.1.6.1 Gas Turbine Enclosure
The gas turbine enclosure protects the gas turbine and other ancillaries located within it from the environment
and provides noise attenuation. It provides a protected working environment and inside lighting for maintenance
with the unit in a nonoperating condition. It protects outside workers from the high component temperatures in an
operating system.
The gas turbine enclosure includes the inlet plenum and the exhaust enclosure. The inlet plenum channels the
air from the inlet air silencer to the gas turbine bellmouth to minimize the inlet air pressure loss. The exhaust enclo-
sure houses the exhaust diffuser/collector that delivers the power turbine exhaust gases to the exhaust transition
duct and silencer. The enclosure serves as a mounting structure for the inlet and exhaust air systems, and as a duct
for passing secondary air over the gas turbine for cooling. The enclosure also provides a limited containment vol-
ume for the fire protection system.
2.1.6.2 Gas Turbine Enclosure Secondary Air System
The gas turbine enclosure secondary air system provides cooling air flow throughout the length of the enclo-
sure. Air is drawn into the intake assembly, through the silencer chamber, and blown into the gas turbine enclosure.
It passes along the length of the gas generator and power turbine, around the exhaust collector and up into the
exhaust air silencer, where it joins with the gas turbine exhaust.
A pair of alternating current motor driven fans are mounted low on the front bulkhead of the enclosure to take
in atmospheric air through a louvered opening mesh screen and a silencer module.
If a gas turbine enclosure fire is detected, the fans are deenergized and a pair of guillotine-type gates are
released to block the inlet air supply ports.
2.1.6.3 Inlet Air Filtration
The primarily inlet filtration system has a filtration efficiency of 99.7 percent for particles down to 5 microns,
and 95 percent for particles down to 2 microns. Clean elements flow about 86 kg (190 lb) of air per second with a
maximum pressure drop of approximately 1 in. (2.54 cm) of water.
The filter house is constructed of mild steel and rests atop (with its flange bolted to) the inlet air silencer mod-
ule. Air enters both endwalls of the filter house. Each endwall is configured to have filter elements and 6,000 half -
sized elements in individual galvanized frames.
The high-efficiency, replaceable fiberglass filters are preceded by a prefilter. The two filters are kept together
by a pair of extended spring clips.
Filter elements can be replaced from inside the filter house. The doors can be accessed from a full-length exter-
nal catwalk with safety railing and fixed-end ladder.
2.1.6.4 Inlet Air Silencer
The inlet air silencer module attenuates noise over a broad band of frequencies.
Module walls, configured for sound attenuation, consist of equally-spaced, transversely-mounted vertical
acoustic baffles. The baffles, which have a 16-gage perforated galvanized steel shell and are filled with 3 in. of
mineral wool, may be removed for servicing.
2-23
2.1.6.5 Intercooler Description

Two types of heat exchange systems were investigated:
1. Water-cooled intercoolers with heat rejection to a cooling tower
2. Air-cooled intercoolers.
The cost of a heat exchanger is a function of the heat transfer surface area and the materials of construction.
The amount and type of heat transfer of a surface used impacts:
• The pressure drop of the fluid streams passing through the exchanger
• The temperature differences between these streams.
For pressure drop optimization, engine air side intercooler pressures drops of 3 percent, 4.5 percent, and 6 per-
cent were investigated. The pinch temperature difference between the cooling medium and the intercooler outlet
(or the HPC inlet) was held constant for each of the intercooler pressure drop cases studied. As expected, the inter-
cooler installed costs were the highest for the 3 percent pressure drop case, since this configuration contained the
most surface area and required more steel and concrete to complete the installation than did the others. However,
the corresponding engine output for this 3 percent case was significantly greater than that of the other two cases
studied. This performance improvement more than offset the increases in associated installed costs resulting in the
lowest dollars/kW value of all the cases investigated.
At the completion of the pressure drop investigation, an HPC inlet temperature optimization study was under-
taken. The pinch temperature difference between the cooling medium and the intercooler outlet was varied, and the
resulting impact on engine performance was calculated. For this study, three HPC inlet temperatures of 31°C
(88°F), 46°C (115°F), and 60°C (140°F) were investigated, while holding the intercooler system pressure drop con-
stant at 3 percent for each of these study cases. As expected, the intercooler installed costs were the highest for the
31°C (88°F) HPC inlet case. The engine output attributable to the lowest HPC inlet temperature case was signifi-
cantly better than that for the other cases. As before, the configuration which produced the most power also had the
highest installed capital cost. Once again, the configuration which produced the most power also resulted in the
lowest installed cost per kW.
Figure 2-21 through Figure 2-24 show isometric and plan views of the plant layout with water-cooled and air-
cooled intercoolers.
2-24
Figure 2-21. Isometric View of the FT60 IC With Figure 2-22. Plan View of the FT60 IC With a
a Water-Cooled Intercooler Water-Cooled Intercooler
As
shown in these drawings, a simple header design was chosen
to minimize the pressure drop through the supply and return
portions of the piping system. Care should be taken during
detailed design to provide an even distribution of the air flow
through each of the cooler bodies or bays to the interconnect-
ing header. Orifice plates may be have to be located in the
flanges connecting the headers to each cooler inlet to ensure
this. A second option, using a candelabra-header design with
three levels of subheaders, was also investigated for the air-
cooled intercooler case.
This candelabra-header design option does ensure even
engine-air flow distribution to each of the intercooler banks.
However, this design option results in a higher-pressure drop, Figure 2-23. Isometric View of the FT60 IC
due to the need for more fittings and the additional linear feet With an Air-Cooled Intercooler
of pipe required completing the assembly. Table 2-2 presents
the data comparing the pressure drops calculated for these two
header options:
Table 2-2. Data Comparing the Pressure Drops Data Calculated for These Two Header Options
Simple-Header Design Candelabra-Header Design
Pressure Drop Without Turning Vanes (pi) 0.76 2.53
Pressure Drop with Turning Vanes (pi) 0.69 0.74
Therefore, although the candelabra arrangement is technically feasible, it is not the best alternative when its
pressure drop is weighed against the orifice plate option.
2-25
The type of intercooler selected depends upon the availability of cooling water at the chosen site. If sufficient
water is available, the water-cooled intercooler is the preferred choice. Whether a cooling tower is applied or not
depends upon access to sea, river, or other open body of water.
Figure 2-24. Plan View of the FT60 IC With an Air-Cooled Intercooler
2.2 COMBINED CYCLE

A combined cycle version of the FT60 IC was evaluated, using a biphase bottoming cycle. This cycle is well
suited to recover energy from low-temperature exhausts. The biphase combined cycle is shown schematically in
Figure 2-25. Exhaust heat from the gas turbine flows through a superheater section and two-phase economizer,
transferring heat to a once-through working fluid. The two-phase mixture from the economizer flows into a biphase
turbine where the two-phase mixture is expanded in the first stage of the biphase turbine to produce power.
2-26
Figure 2-25. Biphase Combined-Cycle Boasts System Efficiency
Separated steam from the first stage flows through the superheater to the inlet of a steam turbine (in a biphase
turbine, the superheated steam enters the structure through a port downstream of the initial biphase stage). Sepa-
rated water from the first stage of the biphase turbine is flashed in a second biphase stage, producing additional
power and intermediate pressure steam and water. The intermediate pressure steam flows to another inlet port of
the steam turbine. The intermediate pressure hot water is then flashed in the third biphase stage, producing addi-
tional power and low-pressure steam and water. The low-pressure steam flows to the third inlet port of the steam
turbine. The separated water has its pressure increased in an internal diffuser and flows back to the inlet of a feed-
water pump. The mixed-steam flows are expanded in the steam turbine, producing power. Steam leaves the steam
turbine at the lowest pressure of the cycle and forms condensation in the condenser. The condensation pump pres-
surizes the condensation and pumps it to the boiler feed pump. The combined hot water stream at high pressure
flows into the biphase economizer to complete the cycle.
2-27
The following two factors increase the efficiency of the

two-phase bottoming cycle relative to a conventional steam
bottoming cycle: 800
Exhaust
700
• The thermodynamic energy loss due to constant
Temperature — °F
temperature steam vaporization is reduced. 600
Biphase
Figure 2-26 shows the heat transfer from the turbine 500 Bottoming
Cycle
exhaust into the biphase working fluid for the Two-Phase
400 Heat
biphase bottoming cycle. The majority of the heat Super-
Heat Exchanger Economizer
transfer is gas to liquid, with no pinch point. A 300
30% Outlet
Dual Pressure
small pinch point exists at the exit of the biphase 200 Vapor Quality
Steam Bottoming Cycle
economizer as the wet steam-water mixture is gen- 10k 20k 30k 40k
erated. However, this pinch point is much lower Heat Transfer — Btu/Sec
than that realized in a conventional steam bottom- 126760.cdr
ing cycle. Figure 2-26. Heat Transfer for the Biphase

Bottoming Cycle
• Mixing hot water leaving the biphase turbine with
condensate produces regenerative heat of the feedwater that is not available in a conventional bottoming
cycle.
In 1995, P&W commissioned an engineering study of the biphase combined cycle that compared a conven-
tional FT8 gas turbine to a proposed FT4000 gas turbine, which has similar characteristics to the FT8000 IC.6 The
efficiency of the biphase bottoming cycle was improved by 30 percent over an optimized bistage steam turbine bot-
toming cycle. The major reason for this was that minimization of the pinch point loss enables more heat to be
extracted from the gas turbine exhaust. The exhaust temperature reached in the biphase cycle was 82°C (180°F),
versus a minimum of 168°C (335°F) found possible with a two-stage steam bottoming cycle. In the current con-
tract, Douglas Energy and MPPT did a conceptual design of the biphase bottoming cycle.
The performance of a single FT60 IC Block 2 gas turbine and one biphase bottoming cycle was about 200 MW
and 59 percent efficiency. This is considerably higher than any current combined cycle at this small an output.
According to the Gas Turbine World 2000-2001 Handbook, the highest efficiency combined cycle in the 150 to 250
MW size class is 53.7 percent for 2 × 1 combined-cycle 6FAs.
In the economic analysis, a single FT60 IC combined cycle was not as cost effective as the FT60 IC, without
bottoming, at the baseline gas price of $3.90 per 106 Btu. If gas prices become extremely high, ($13.50 was chosen
as an example), combined cycles were the economic choice starting at 2,000 to 2,500 hr/year. Also, one 200 MW
FT60 IC combined cycle was slightly less cost effective than the 590 MW 7FX-2 combined cycle, due to econo-
mies of scale.
2-28
3. CONCLUSION
3.1 MARKET STUDIES

A market study for the FT60 IC was performed by the ADL and by P2 Energy. The market study consisted of
stakeholder interviews, economic analyses, and a market forecast. PWPS provided the attributes of the FT60 IC
and conducted some of the interviews. The economic analyses were discussed earlier in the description of system
attributes. The approach and other key results of the market study are summarized in this section. The high output
of the FT60 IC precludes its application in any market except electric power generation; there is assumed to be no
market demand for mechanical drive or marine propulsion applications.
ADL evaluated the market opportunity for the NGTs being developed at P&W. This market evaluation con-
sisted of:
• Stakeholder interviews
• Economic analysis
• Market forecast.
ADL's Task Our Approach/Actions

• See stakeholder input regarding: • Conducted a mini-workshop with P2 Energy to identify and select
– Their future power generation needs key stakeholders for interviews.
– Importance of GT performance attributes. • ADL designed the interview guides and reference material.
• Jointly with P2 Energy, conducted key stakeholder interviews with
21 large players in the power generation business.
• ADL compiled and analyzed the feedback and its implications for the
NGT development team.
• Determine Market Potential for NGT's • ADL analyzed the business conditions and trends in global and U.S.
– Economic analysis power generation markets to identify the overall size of the market
and opportunities for using gas turbines under development.
– Estimate target market for NGTs
• ADL created three scenarios to identify key drivers and establish a
– Sensitivity analysis top assist the development team.
framework for economic analysis and market forecasts.
• We used ADL's market forecast model to assess the U.S. market
opportunity for NGT's by accounting for planned additions, load
growth, and future plant requirements for the next 20 years.
• ADL estimated global market potential for NGT's. We used insights
from our U.S. analysis to forecast the market potential globally.
The NGT (FT60 IC) is an intercooled gas turbine rated at 170 MW. The design targets for this turbine are sig-
nificantly better than any other turbine on the market today. The targeted efficiency, in simple cycle application, is
greater than 50 percent, while the best comparable turbine on the market is at 38 percent. The commercial avail-
ability date of this NGT is projected to be 2007.
ADL performed their analysis within the framework of the scenarios show in Figure 3-1. The greatest drivers
for the NGT market are load growth and gas prices.
3-1
Market Driver Impact 2005-2020

2001-2005 GT Bubble
+ Overbuilt Bust
Natural Gas Prices ++ 2.8% CAGR 1.2% CAGR
Gas Pipeline Development + Limited Robust
High Coal Prices + Low High
Peak Load Growth +++ 1.3% CAGR 2.5% CAGR
DisGen/EnergyEff. + High Low
Environmental Factors Light Green Dark Green
++
Transmission Capacity Growth Sustained Limited
+

2001-2005 GT Bubble
+ Overbuilt Bust
++ 1.6%
Natural Gas Prices 2.8% CAGR 1.2% CAGR
Baseline Coal Prices + Low High
1.9%
++
+

2001-2005 GT Bubble
+ Overbuilt Bust
Natural Gas Prices ++ 2.8% CAGR 1.2% CAGR
Low Coal Prices + Low High
++
+
126761.cdr
Figure 3-1. Framework of the Scenarios for the ADL Analysis
We interviewed 21 large players in the global power generation market to seek their perspectives on the overall
generation markets and the new turbine(s) under development.1 Most respondents did not distinguish between base
load and intermediate duty plants. Many of the interviewees were merchant plant developers who believed that
capacity factor of the plant will vary with market dynamics. Even though the specific needs of customers vary
between combined cycle owners and those owning peakers, both groups selected the same top three desirable
attributes (availability, reliability, and economics). In general, they all saw a continuation of the trend towards
building large combined cycle projects based on 7F and 501G technology. However, they did not see any need for
even larger gas turbines. They seem to prefer a relatively smaller peaking unit size equivalent to 100 MW installed
in banks of four or eight at a location. Presently, the 7EA seems to be the most popular choice for a peaker.
Leading energy trading companies tend to use peaking NGTs’ Relative Attractiveness for Different Applications
units as a financial hedge. They view them as a conversion Intermediate Base
Peaking
device to pump electricity into the grid at a very short Load Load
notice. They prefer a large peaking unit equivalent to 175 NGT SC
MW. Reliability is a bigger concern for this group. NGT CC
Most respondents considered economic factors as being Very Attractive Attractive Partially Attractive
critical to succeed in the merchant power market. However,
Marginally Attractive Not Attractive
even operational considerations or noneconomic factors
126762.cdr
have a quantifiable impact on the economics (dollars/MW/ Figure 3-2. Comparative Economic Analysis
hr), thereby blurring the distinction between the two.
Comparative economic analysis shows that NGT simple
cycle will be attractive for both peaking and intermediate duty applications. The combined cycle, however, is only
partially attractive for baseload applications (Figure 3-2).
The market for new generating capacity in the U.S. was estimated based on supply-demand growth projections.
Overbuilding the retirement over the years is also taken into account (Figure 3-3).
1
Large gas turbines (greater than 60 MW) constitute more than 85 percent of the total turbines sold in the U.S. during the last 10 years.
Large turbines are also expected to maintain the percentage in the future.
3-2
Lower U.S. 48 States

Summer Capacity Growth Projection -- Base Case Scenario
1200
and
em
e et D in
to M arg
1100
p a city erve M
Ca Re s
red 15% nd
Over Supply qui De
ma
Re and e t
e t
1000 to M arke
pa city mic M
Ca cono
ted ew
d n th —
E
jec th e
900 L Pro Grow q u i r
r o w
AD acity Re G )
p and .9%
Ca D em R = 1
oad , CAG
dL
800 j e cte enario
Pro e Sc
s
e Ca
B a s
(
700
600
1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019
* Assuming conservatively that 0.9% of the 1999 installed capacity would be
retired each year between 2005 and 2020.
126763.cdr
Figure 3-3. Market for New Generating Capacity in the U.S.
About 370 GW of new generating capacity Total Required New Generation Additions in the U.S.*
would be required in the U.S. between 2005 and During (2005-2020) in “Thousand MW”
2020 to keep pace with demand growth and to off-
set estimated plant retirements. It is widely
believed that gas turbine-based simple-cycle and 524
combined-cycle powerplants will account for about 372
236
90 percent of the new generating capacity additions
in each scenario (Figure 3-4).
1.3% 1.9% 2.5%
The global market for NGTs is estimated to be Annual Demand Growth Rate
over $30 billion and is almost equally divided
between the U.S. and the rest of the world *The U.S. Lower 48 States.
126764.cdr
(Table 3-1 and Table 3-2). Figure 3-4. Generation Additions in the U.S.
Table 3-1. NGT Market Potential in U.S. (2007 to 2020)

Total Market for Gas Turbines NGT Projected Market
Percent Billions of
Scenario MW Share MW Dollars
Low growth at 1.3 percent CAGR 213,000 22.3% 47,500 $10.5
Base case at 1.9 percent CAGR 335,000 19.9% 66,500 $14.6
High Growth at 2.5 percent CAGR 420,000 19.5% 81,900 $18.0
OEM price used in computations above dollars/kW: $220.0
3-3
Table 3-2. NGT Market Potential in the Rest of World (2005 to 2020)1
Total Market for Gas Turbines NGT Projected Market
Percent Billions of
Scenario MW Share MW Dollars
Rest of world at 2.7 percent CAGR 476,000 14.9% 71,100 $15.6
1
We have not made any allowances for: a) any delayed adoption of new technology, and b) plant retirements in the rest
of the world.
NGTs could generate total sales of more than $15 billion over the planning horizon based on its $220/kW price.
While significant additional sales will result from services and spare parts, their impact is not included here
(Figure 3-5).
Total Sales Forecast Average Annual Sales
(in ‘000 MW) (in MW)
82
67
5900
48 4800
3400
1.3% 1.9% 2.5%

1.3% 1.9% 2.5%
CAGR per Low/Base/High Scenario CAGR per Low/Base/High Scenario
126765.cdr
Figure 3-5. NGT Sales Forecasts
The attributes of the FT60 IC, combined with the market forecast above, will generate a considerable amount
of public benefits, as indicated in Figure 3-6. Adoption of NGTs will lead to substantial economic, environmental,
and societal benefits.
Reduced Air
Emissions
(CO2, Nox , SOx )
Lower Primary
Energy Conservation of
Consumption Natural Resources,
e.g., Land, Water
Public Benefits
Fuel Cost Savings from NGTs
Improved U.S.
Competitiveness by
Increasing the
Lowered Cost of Export Potential
Electricity
Improved System
Reliability
126766.cdr
Figure 3-6. Benefits of Adopting NGTs
3-4
NGTs “S” Curve of

Addressable Total Primary Fuel Cost
Technology
Market Displaced (BTU) Savings
Adoption
Potential
Characteristics
of Technologies Emissions
Emissions Reductions
Displaced*
Factors (CO2, Nox , Sox )
NGTs
Characteristics
Note: * Based on published industry data and ADL Research

126767.cdr
Figure 3-7. Savings Were Generated Using Market Projections
Energy savings, electricity cost savings, fuel

savings, and emissions reductions were calculated Cumulative Installed Capacity of NGTs in the USA
using the base case NGT market projections

70000
(Figure 3-7). The public benefit calculations are
Replacing Existing Units
based on our market forecast assumptions and pro- 60000
Replacing Future Units
jections for NGTs. NGTs are projected to displace
Total Installation — MW
Total
50000
some old generating plants and future GT plants
40000
based on E&F technology. Existing units are those
oil-fired and gas-fired, intermediate duty, steam-tur- 30000
bine plants that would be displaced by NGTs. 20000

Among the total capacity of this segment, gas-fired
10000
plants account for about 75 percent of the total,
1
whereas oil-fired account for the other 25 percent. 0
2005 2010 2015 2020
Heavy oil prices are used to represent prices of oil
as they account for more than 95 percent of the total 126768.cdr
fuel oil consumption. Future plants that would be Figure 3-8. Capacity of NGTs
displaced by NGTs are mostly composed of gas tur-
bine simple-cycle powerplants for peaking and intermediate load applications (Figure 3-8).
Economic benefits are dependent on the future U.S. Fuel Price Projections, 2005 - 2020
fuel oil prices of gas and oil. Gas prices were taken
5.0
from our study and oil prices forecasted by EIA
Dollars/MMBTU, in 2000 Dollar
(Figure 3-9). 4.5 Natural Gas

Fuel Price —
The efficiencies and emission factors for NGTs Heavy Oil*

4.0
and generation technologies are shown in
Figure 3-10.
3.5
Annual fuel saving from the adoption of NGTs
would reach almost 190 trillion Btu in 2020, which 3.0
2005 2010 2015 2020
is enough to fuel a 3,000 MW plant for 1 year Note: 1. Oil price is based on EIA Annual Energy Outlook 2001; 2. Gas price is ADL
(Figure 3-11). projection based on RDI and EIAforecast.
126769.cdr
Figure 3-9. U.S. Fuel Price Projections
1
Percentage assumptions are based on RDI PowerDAT.
3-5
Existing Technology that Would Future Technology

Be Displaced by NGTs that Would Be NGTs
Gas -Fired Oil -Fired Displaced by NGTs
LHV Efficiency 35% 35% 39% 50.2%
CO2 Emission Factor

(lbs/MWh) 1151 1463 1030 800
SO2 Emission Factor

0.006 8.4 0.005 0.004
(lbs/MWh)
NOx Emission Factor
1.8 3.7 0.31 0.24
(lbs/MWh)
Source:
1. EPA E-GRID Database
2. New Energy Market Power
3. ADL Study
126770.cdr
Figure 3-10. Efficiencies, Emission Factors, and Displaced Generation
Technologies
Annual NGTs Fuel Saving in the USA due to Adoption of NGTs, in Trillion BTU
200 From Replacing Possible New Units

From Replacing Existing Units **
200
Oil
180 Total Fuel Saving 180
Primary Fuel Saving Trillion BTU
Primary Fuel Saving Trillion BTU
Gas
160 160 Total
140 140
120 120
100 100
80 80
60 60
40 40
20 20
2005 2010 2015 2020
0 0
2005 2010 2015 2020 2005 2010 2015 2020
** Existing units that would be displaced are mainly oil-fired or gas-fired steam turbine power plant.
126771.cdr
Figure 3-11. Fuel Savings
3-6
Annual NGTs Fuel Cost Saving in the USA
900
800
Annual Fuel Cost Saving —

700
Total Saving
600
Millions Dollars
500 Gas
400
300
200
100 Oil
0
2005 2010 2015 2020
126772.cdr
Figure 3-12. Fuel Cost Reduction
Annual fuel cost reduction from the adoption of NGTs would reach 800 million dollars in year 2020 (in year
2000 dollars) (Figure 3-12).
Adoption of NGTs will lead to substantial reduction in CO2 Annual CO2 Savings
emissions (Figure 3-13). Annual reduction in CO2 emissions
would be over 26 million tons per year in 2020. This is equiva- 30,000
lent to: 25,000
• Total emission of 14 × 300 MW coal-fired steam-turbine 20,000
powerplant (assuming a 75 percent capacity factor
15,000
• Total emissions of 50 × 300 MW gas-fired steam-turbine
10,000
powerplant (assuming a 40 percent capacity factor).
5,000
The adoption of NGTs will lead to a reduction in NOx and
0
SOx emissions as well (Figure 3-14). The majority of the SO2 2005 2010 2015 2020
and NOx savings is from the displacement of existing gas-fired
126773.cdr
and oil-fired steam turbine plants. In 2020, the annual NOx and
Figure 3-13. Reduction in CO2 Emissions
SO2 savings are estimated to be 21,000 tons and 14,000 tons,
respectively. The NOx emission savings are equivalent to the
total emissions of approximately 3 to 6 units of 300 MW coal-fired powerplant. The SOx emissions savings are
equivalent to the total emission of approximately 1 to 2 units of 300 MW coal-fired powerplant. NOx and SO2 are
not significant as that of CO2 because NGTs would be primarily used to displace the conventional gas turbine tech-
nology, which already has limited emissions of NOx and SO2.
The cumulative energy and emissions savings could be substantial especially in the later years when NGTs
become widely adopted (Figure 3-15).
The use of all types of gas turbines will reduce the need for land and water resources as compared to the steam
plants they might displace (Figure 3-16).
Total sales of NGTs in the rest of the world is projected to be $15 billion, about the same as in the U.S.
(Figure 3-17).
U.S. turbine manufacturers would capture a significant portion of the global market of NGTs, as well as other
service business opportunities. Assuming U.S. manufacturers could gain 60 percent of the export of the rest of the
3-7
world market, this would result in $9 billion of U.S. exports. U.S. companies will also earn revenues from ancillary
services, spare parts, and other business related to NGTs. As a result, the competitiveness of the U.S. turbine manu-
facturers will be improved. In addition, export of NGTs would create new jobs in the U.S.
There are significant other benefits realized from the adoption of NGTs, although they may be difficult to quan-
tify. NGTs would help the U.S. address the global warming issue more easily. There will be a significant amount of
CO2 reductions in the U.S. resulting from the use of NGTs (as calculated previously). The U.S. may also benefit
from the credits of CO2 reductions from other countries use of NGTs. NGTs would also have other environmental
benefits such as reduced particulate emissions. U.S. customers may benefit from reduced retail electricity prices
resulting from low-cost power generated by NGTs. The NGTs’ higher performance and availability would also
improve the system reliability of equipment owners and the power grid.
Annual NOx and SO2 Savings
25
20
NOx
Thousand Tons
15
SO2
10
0
2005 2010 2015 2020
126774.cdr
Figure 3-14. Reduction of Other Emissions
Cumulative Savings from 2007
2010 2015 2020

Primary Energy
44 350 1,100
(Trillion BTU)
Fuel Costs Savings
165 1,400 4,640
(MM 2000$)
CO2
5,020 45,000 148,000
(Thousand Tons)
SO2
7 50 100
(Thousand Tons)
NOx
7 52 133
(Thousand Tons)
NGTs Installation
6,300 33,000 71,000
(MW)
126859.cdr
Figure 3-15. Cumulative Energy and Emissions Savings
3-8
Percent
Gas Turbine Steam Plant*
Reduction
Land (Acres) 5–15 25–50 60%–90%
Service and Plant Water

(mgd) 1–2 0.5–1
Cooling Tower Makeup
Water
Water (mgd) 0–8 12–15
Waste Water Discharge
(mgd) 1–8 8–14
Overall (mgd) 2–18 20–30 30%–90%
*: Gas and Oil (100MW-300 MW).

Source: ADL Estimate
126775.cdr
Figure 3-16. Advantages of NGTs Over Steam Plants
NGTs Total Sales in the Rest of the World (in Million $2000), 2007 - 2020
Total Sales: $15 Billion

Annual Average: $1.1 Billion
3,669
2,954
1,993
1,615 1,463
910 1,089
836 638 466
Africa China Japan Middle Other Asia Canada Mexico Central & Eastern Western
East & Pacific South Europe & Europe
America Former
U.S.S.R.
* Assuming a Price of $220/KW
126776.cdr
Figure 3-17. Projected Sales of NGTs Worldwide
3-9
3.2 PUBLIC BENEFITS

3.2.1 The System’s Ability to Serve Multiple Uses
KraftWork Systems carried out analyses to identify the NGT engine system's ability to serve multiple uses. The
majority of these uses are based on employing the NGT system in conjunction with a coal-fired plant, or on using
coal as the system fuel. Also identified was the ability of the NGT system to serve as the basis of an advanced per-
formance cycle, the HAT cycle.The HAT cycle is also is used with coal gasification in an IGHAT. The NGT sys-
tems to be identified are:
• Feedwater heating retrofit to an existing coal-fired steam plant, which could supply both heat and peaking
power (Block 2 engine)
• Repowering of an older coal-fired plant (Block 2 engine)
• Gas-fired HAT cycle (Block 1 and 2 engines); integrated gasification HAT (Block 1 and 2 engines).
In addition, results from a recent study funded by DOE (contract DE-AC22-95PC95144) concerning applica-
tion of the intercooled aeroderivative engine technology in an indirectly fired gas turbine will be included to round
out the potential cycle/fuel flexibility of this technology.
3.2.1.1 New Coal-Fired Plant with Feedwater Heating
The NGT can enhance the perfor- 100 Gas Turbine MW
mance and flexibility of new coal-fueled from Natural Gas
powerplants. For example, in the near 75
Load Percent
term, the NGT could be used with con- Steam Turbine MW

ventional coal-fired steam plants. The from Natural Gas
50
heat from the intercooler and the exhaust Steam Turbine MW
of the natural gas fired NGT can be used 25 from Coal
for feedwater preheating in coal fired
steam plants. This increases the output 0
and efficiency of the coal-fired plant, 0 4 8 12 16 20 24
Typical Daily Load Pattern
because steam does not need to be 126295.cdr
extracted from the steam turbine for Figure 3-18. Typical Daily Load Factor for Feedwater Heating Plant
feedwater preheating. The gas turbine is
operated on natural gas fuel during times of peak demand. The plant operator sells power only into the base load
market with the coal plant when the gas turbine is not operating and sells both the power from the gas turbine and
the increased output of the steam plant in the on-peak market when he operates the gas turbine. If needed, the
remaining heat from the gas turbine exhaust can be used to heat water for process or district heating. This equip-
ment can be added to existing coal-fired plants as well as new plants. The cycle arrangement and the plant output
during a 24-hour period are shown in Figure 3-18 and Figure 3-19.
As an example of this approach, the steam system shown in Figure 3-20 has been investigated. This basic
steam plant operates at 3500 psi/537ºC (1000ºF). With extraction for seven feedwater heater stages, the heat rate is
9,077 Btu/kW/hr and the output is slightly over 800 MW. The plant is designed to accept full throttle flow during
periods when peak power is needed. During this peak demand period, the feed water is heated by the exhaust from
five NGT engines, which supply a further 842 MW of power (Figure 3-21).
The performance of this feedwater-heating cycle is quite impressive. The output of the system has been
increased from 800 MW to 1,850 MW. Similarly, the heat rate has been reduced from 9,077 Btu/kW/hr to 7,155
Btu/kW/hr, a decrease of over 21 percent. At full peak output, gas supplies 47 percent of the system energy. The
incremental gas efficiency in this system is 57 percent (steam MW from gas and GT MW)/gas energy lower heating
value [LHV]). While analyses were performed only for the case when all of the extraction steam is expanded, it is
possible to operate with partial extraction and use only selected gas turbines to obtain the desired power output.
This gives great operating flexibility to the concept. In Figure 3-21, the extraction heaters and the necessary valves
and piping have been omitted for clarity.
3-10
Conventional Steam Plant

ESP
Air
Desulfurization
Air
Preheater
DeNOx
Coal
HRSG
Air
Natural Gas
Gas Turbine Feed Water Heating

126777.cdr
Figure 3-19. Diagram of Feedwater Heating Cycle
P=644
T=1000 P = Pressure - psi
M=4793543 T = Temperature - °F
P=715 Reheater
T=654 M = Flow - lb/h
P=75.3
M=4793543 T=570 Heat Rate = 8865 Btu/kWh
M=3942150
Boiler
P=3675
T=1000 HPT IPT LPT 800,834 kW
M=5523498
M=3014537
P=4612 1.5in.
T=502 Hg
M=5523498 M=373811 M=373811
to Feed Pump from Feed Pump
T=101
M=3703976
T=280 Q=102229738
Q=603718331 Btu/hr
Q=1010192781 T=208 T=169
T=400
T=370
M=5523498 T=245
Q=140009159
Q=1717808438 Q=145195859
Q=130379955
126296.cdr
Figure 3-20. Base-Load Supercritical Steam Plant
3-11
P=644
T=1000 Before After
P=715 M=5523498 Net Power, MW = 800
T=654 Reheater
P=75.3 1849.4
M=5523498 T=570 Net Efficiency, % (HHV) = 38.9 47.7
M=5523498 Incremental Gas Eff. % = 57
Boiler
P=1815
T=1000 HPT IPT LPT 1040 MW
M=1,053,000
P=4612 T=88
T=502 P = Pressure - psi
T = Temperature - °F
M=5523498
M = Flow - lb/h
FWH0
FWH2 FWH1
From GTIC
To FWH0 To FWH0 To FWH0 To FWH0 To FWH0
168.5 MW 168.5 MW 168.5 MW 168.5 MW 168.5 MW

126297.cdr
Figure 3-21. Feedwater Heating for Peak Power
3.2.1.2 Repowering of a Coal-Fired Powerplant

In the next two decades, it is estimated that over 400,000 MW of capacity addition will be needed to meet
increased electrical demand and to replace aging units. This includes approximately 73,000 MW of current coal-
fired power scheduled to be retired during this period. Many of the plants currently scheduled for retirement could
be successfully repowered using NGT technology, thereby reducing the required capital investment in generating
capacity while also reducing energy use and environmental impact including greenhouse gases.
The repowering of aging steam stations with modern gas turbines may now present a better opportunity to pro-
vide the additional power that is required for the future. The ability to use existing facilities and infrastructure and,
in many cases, use locations near to load centers that already have needed environmental permitting makes repow-
ering attractive. A concept definition study was performed to identify the alternative gas turbine configurations to
be used for repowering. The intercooled aeroderivative-type engine was selected to allow the widest flexibility in
operation and also to minimize the footprint of any repowering additions. A specific steam powerplant was
selected to be the basis for a repowering study to estimate potential performance using an actual steam power sys-
tem.
The steam system selected for repowering is shown in Figure 3-22. As can be seen, the system uses modest
steam conditions of 1,815 psi/537ºC (1000ºF). There are a number of approaches to repowering, the most common
being to replace the existing boiler by a heat recovery steam generator (HRSG), which uses the gas turbine exhaust
as the heat source. This means the entire plant capacity is now gas fueled. This approach would not be suitable for
the NGT engine as the exhaust temperature is too low (the equivalent of 426ºC [800ºF]). The repowering method
that makes the best use of the low-temperature exhaust would use this heat to replace the feedwater heating nor-
mally done by steam extracted from various points in the steam cycle; for example, in the steam system selected for
analysis, there are seven extraction points. While this approach, shown in Figure 3-23, allows retention of the orig-
inal steam generator, there would be significant modifications necessary to the steam turbine.
3-12
P=430
T=1000 P = pressure - psi
P=478 M=1053000 T = temp. - F
T=654 Reheater M = flow - lb/h
M=1053000
P=75.3
T=570 Heat Rate = 9250 Btu/kWh
M=678410
Boiler
P=1815
T=1000 HPT IPT LPT 151.6 MW
M=1,053,000
M=443460
P=2440
T=492
M=1,074,000
T= 81
Q= 58166448
Q= 47688264 Btu/hr
T= 211 T= 159
T=452 T= 319 T= 259
T=372
Q= 39106256
Q= 91509912 Q= 48554196
Q= 42210558
126660.cdr
Figure 3-22. Schematic of Steam System
P=430
T=1000 P = pressure - psi
P=478 T = temp. - F
M=1053000
T=654 Reheater M = flow - lb/h
M=1053000 P=75.3
T=588
M=1053000
Boiler
P=1815
T=1000 HPT IPT LPT 198.6 MW
M=1,053,000
T = 81
P=2440
FWH0
T=492
M=1,074,000 FWH2 FWH1
From GT IC
To FWH0 Before After

Net Power, MW = 159.6 354
Net Efficiency, % (HHV) = 36.9 44.8
Incremental Gas Eff. % (LHV) = 51.3
168.5 MW
Coal/Gas Energy = 52/48
126661.cdr
Figure 3-23. Schematic of Repowered Plant
3-13
As designed, the flow of the steam passing through the turbine sections is reduced at each extraction point until
the flow from the LPT to the condenser is only a fraction of the throttle flow into the HPT. When the extractions are
eliminated, modifications to the HPT and LPT will be needed to accommodate the increased flow. Additionally, the
generator will have to upgraded or replaced by one of a higher capacity. When the full throttle flow is expanded,
the steam turbine output is 198.6 MW, nearly 40 MW higher than previously. One NGT is needed to supply the
feedwater heating. Three heat exchangers are used: a low-pressure and a high-pressure feedwater heater and the
engine intercooler. To simplify the Figure 3-23 schematic, the water/steam circuits are not shown in full detail. The
repowered system now has a net output of 354 MW and a heat rate, based on the high heat value (HHV) of the
combined coal and gas fuel, of 7,618 Btu/kW/hr, a 17 percent reduction. Coal furnishes 52 percent of the cycle
heat. The emissions of CO2/kW/hr will be essentially halved compared to the original plant (due to increased effi-
ciency and use of gas for nearly one-half the energy input). The gas fuel is used at an incremental efficiency (addi-
tional power/gas energy) of 51.3 percent (LHV).
In cases where it is not economically feasible to upgrade the steam turbine and generator for increased output,
the replacement of the extraction by gas turbine heat recovery could result in savings in coal flow of 20 to 25 per-
cent (the feed water represents about 25 percent of the heat input, depending on the actual cycle parameters.) This
also represents about a 10 to 12 percent reduction in CO2 emissions compared to the all-coal case.
3.2.1.3 Humid Air Turbine Cycle
In addition to being a worthwhile product on its own, NGT is the first step to an HAT cycle (Figure 3-24). In
the HAT cycle, the compressor discharge air is humidified in a tower, recuperated, and sent to the combustor. This
results in a much higher mass flow and a higher efficiency. The higher mass flow would require a redesign of the
turbine gaspath, but would retain the much more expensive compressor gaspath. The HAT cycle offers equal or
better performance than a combined cycle in a package that would also be less expensive, due to the elimination of
the HRSG, steam turbine-generator, cooling tower, and condenser. This more than compensates for the cost of the
saturator, recuperator, and economizer in the HAT cycle. At the Block 1 design point, temperature (CET = 1330ºC
[2426ºF]), the NGT HAT would have an output of 242.5 MW and an efficiency of 54.3 percent (LHV). At the
Block 2 combustor exit temperature of 1510ºC (2750ºF), the output is 288.1 MW and the efficiency is 57.0 percent
(LHV). Because the moisture addition suppresses NOx emissions, higher turbine temperatures are possible and at
3-14
1704ºC (3100ºF), the NGT HAT has an estimated output of approximately 321 MW and an efficiency of 60.4 per-
cent (LHV).
3.2.1.4 Indirectly Heated Gas Turbine Cycles Using Coal
The following paragraphs describe power systems that use advanced technology coal-fired air heaters to heat
the discharge air from the compressor of aeroderivative gas turbines using technology similar to the NGT engines.
These systems were examined as part of DOE-funded (contract DE-AC22-95PC95144) Engineering Development
of Coal-Fired High-Performance Power Systems (HIPPS), Phases II and III. Its inclusion here is to identify the
potential capability of the NGT-type turbine to use coal as a major fuel source.
In the HIPPS application, (Figure 3-25) the compressor discharge air is heated to 954ºC (1750ºF) or higher in
the high temperature advanced furnace (HITAF), first in a convective heater and then in a specially designed radia-
tor. From there, it goes to a special duct burner where gas is burned to raise the temperature to the desired level
(equivalent to 1371ºC [2500ºF] and higher). The gas turbine exhaust is then sent to a HRSG. When used in the
basic HIPPS configuration, essentially a combined cycle, with technology similar to the Block 1 NGT, the output of
the system is nominally 275 MW with an overall efficiency, based on the HHV of the coal and gas, of 48.8 percent.
Coal provides 65 percent of the system energy input. An advanced system using gas turbine technology similar to
the Block 2 NGT combined with an advanced ultracritical steam system would bring the overall cycle efficiency to
53 percent (combined HHV).
As in the case of the gas-fired NGT, the HIPPS can benefit from the HAT-cycle configuration. An NGT-tech-
nology based HIPPS/HAT is shown in Figure 3-26. In this configuration, which uses a coal-fired reheater and
steam injection before the power turbine, the HAT output is over 395 MW and the estimated efficiency is 54.1 per-
cent (combined HHV) while burning 62 percent coal.
Intercooler
To Saturator
Gas
Aftercooler
Regenerator
To Saturator
Satur ato r
Saturator
Air
Water Preheater
Saturated Air
126662.cdr
Figure 3-24. Simplified Diagram of HAT Cycle
3-15
Intercooler
Condensate To HRSG/Feedheater Air/Combustion Products
1370 C/2500 F Water/Steam
From IC
Air
125 MW
Aero-Derivative Gas Turbine

Gas
357 C/675 F
HRSG
Coal
927 C/1700 F
RAH
BH
CAH FGD
To IC
HIPPS EFF. = 48.5% HITAF
Coal/Gas = 65%/35% Condenser
566C/1050 F/166 bar/2400 psi

120 MW
Steam Turbine
566 C/1050 F/28 bar/400 psi
126663.cdr
Figure 3-25. HIPPS With NGT-Type Gas Turbine
3-16
Intercooler
To Saturator Air/Saturated Air
t 2780°F Water
Aero-Derivative
Gas Turbine Air
395MW
To Saturator
HIPPS EFF. = 54.1%
Gas Coal/Gas = 62%/38%
Saturator Recuperator
Air
1975°F
Air Preheater
1975°F
Coal
To Saturator
RAH
Air Preheater
BH
CAH FGD
ToSaturator
HITAF
955°F
126664.cdr
Figure 3-26. A HIPPS/HAT System
3.2.1.5 Integrated Coal Gasification Humid Air Turbine Cycles

When integrated with a coal gasification system, the NGT HAT has the further advantage that it can be devel-
oped to achieve the same attributes as the integrated combined-cycle powerplant, with the additional advantage of
lower capital cost. An IGHAT configuration is shown in Figure 3-27.
Based on studies done in cooperation with Texaco, Fluor Daniel, and EPRI for a HAT cycle based on an aero-
derivative version of the PW4000 aircraft engine, estimates indicate an IGHAT would have an efficiency compara-
ble to that of a heavy-frame engine integrated combined-cycle powerplant. The IGHAT has the potential for lower
cost than the integrated combined-cycle powerplant because it could operate without a steam turbine system and its
accompanying equipment.
Performance for Block 1 and Block 2 versions is given in Table 3-3. The IGHAT analyzed used a Texaco high-
efficiency quench (HEQ) gasifier. The reason for using this gasifier was to give a basis of comparison to the H
engine integrated combined-cycle powerplant performance given in the literature (43 percent HHV). As can be
seen in the following table, the estimated efficiencies of both the Block 1 and Block 2 IGHAT systems are higher
than those given for the H-class integrated combined-cycle powerplant, even though they use some steam-system
components to recover the lower grade heat in the HEQ system. Based on prior analyses of integrated combined-
cycle powerplants and HAT cycles, it is expected that the IGHAT would benefit more than an integrated combined-
cycle powerplant when employed in conjunction with a dry-feed gasifier coupled to a high temperature gas cleanup
system. An IGHAT with a NGT operating with combustor exit temperature of 1676ºC (3050ºF) was also analyzed.
The powerplant output was 479 MW at an estimated efficiency of 50.1 percent (HHV).
3-17
Table 3-3. FT60 IGHAT Performance Estimates (Texaco HEQ Gasifier)

Block 1 Block 2
HAT Power (MW) 299 359.9
Steam Power (MW) 54.6 69.6
Expander Power (MW) 4.3 5
Total Gross Power (MW) 357.9 434.5
Auxiliary Power (MW) 43.6 51.4
Net Power (MW) 314 383.1
Overall Efficiency Percent (HHV) 44.1 46
Air Compressors
Air
Coal
Slurry NGT
Prep Air
Air Separation
Unit
~
Cooling
O2 Compressors
Intercooler
Recuperator
Economizer
Scrubbed
Gas
Saturator
Texaco
Gasifier
Steam BFW
Selexol
Unit
Raw Water
Water
Treatment Deaerator
O2/Air/Combustion Prduct
Water/Steam/Humidified Air
Syn Gas
126665.cdr
Figure 3-27. IGHAT Schematic
3-18
3.2.2 Support of Vision 21

KraftWork Systems carried out analyses to identify how the NGT engine system could be integrated into
Vision 21 plants. Vision 21 identifies the DOE plan for advanced coal-and gas-based power systems for introduc-
tion in the 2015 time frame. The Vision 21 goals include electrical efficiencies of 60 percent (HHV) on coal and 75
percent (LHV) on gas. These very high efficiencies are to be attained with systems having near zero emissions of
pollutants including greenhouse gases. Program funding limitations only allowed consideration of gas-fired sys-
tems. Comments of the public benefits that could accrue with the NGT-based Vision 21 plants are also included.
3.2.3 Next Generation Turbine/Fuel Cell Hybrids

As previously described, the highest efficiencies for NGT systems were those of the gas-fired NGT HAT, which
could reach efficiencies equivalent to 58 to 60 percent using Block 2 technology and beyond. To reach the lofty
DOE goal of 75 percent (LHV), use of new technology in areas other than the NGT would have to be investigated.
The fuel cell/gas turbine hybrid offers this opportunity.
At this time, solid oxide fuel cells (SOFC) have been demonstrated in the greater than 200 kW size range. The
SOFC operates at high temperature, approximately 1000ºC (1800ºF), and has been pressurized to as much as 1.01
Mpa (10 atm). This technology can be used in a new type of hybrid cycle in which the SOFC tops the gas turbine.
This hybrid cycle is thermodynamically similar to a gas turbine/steam turbine combined cycle in that the highest
cycle efficiency is attained when all of the fuel into the system is used first in the topping portion. The SOFC, like
other fuel cells, converts hydrogen to water while producing electricity. Since the fuel available is methane, a
reformer, which converts much of the methane to carbon monoxide and hydrogen, is necessary. The fuel cell is
then fed a fuel mixture of hydrogen, carbon monoxide, and unreformed methane. The SOFC does not fully use all
of the input fuel; typical utilization factors are in the 70 to 80 percent range. Thus, the SOFC exhaust contains
small fractions of the input fuel. This mixture can be combusted to raise the temperature of the exhaust from the
1000ºC (1800ºF) range to values up to 1204ºC (2200ºF). This temperature is well below the combustor exit temper-
ature for the Block 1 NGT. The hybrid system, however, does benefit from the higher compressor and turbine effi-
ciencies of the NGT and also from the higher power density resulting from intercooling. An NGT-based hybrid is
shown in Figure 3-28. The performance is outstanding, with an estimated efficiency of 77.8 percent (LHV).
High Pressure GT/SOFC Hybrid
LPC HPC HPT LPT Gen
Intercooler
Total Gross Power - 654 MW DC

SOFC Power - 470.4 MW AC
GT Power - 183.6 MW
Efficiency(LHV) - 77.9%
Saturator
GT Working Fluid From Water

Water Treatment
Fuel
126666.cdr
Figure 3-28. NGT/SOFC Hybrid
3-19
The combustor exit temperature for this system is 1167ºC (2134ºF), a value far below the Block 1 value. The
use of the ceramic 1st-stage vanes means that the turbine can run essentially uncooled in hybrid mode. Although
the turbine temperature is down, the output is still over 180 MW, mostly do to the additional flow of steam from the
fuel cell. It is, in effect, a steam injected engine. Operation at near 6.08 Mpa (60 atm) and, at a size of 470 MW, is
far beyond current SOFC planning. Nonetheless, the very high efficiency of the hybrid makes this approach a very
desirable subject for further research and development.
3.2.4 Comments on Public Benefits of the Vision 21 Application
A system with an efficiency of over 77 percent could offer a number of benefits to the general public. Even
with higher cost gas, versus lower cost coal, the cost of electricity (COE) has an opportunity to be reasonable, if not
actually more attractive. The cost of the SOFC has yet to be determined. The DOE goal is for fuel cell costs to be
below $1,000/kW. If this ambitious goal can be met, then the COE would indeed be significantly lower than most
coal-fired alternatives.
This potential for low COE is accompanied by an extremely low pollution profile. The only source of sulfur is
the natural gas. It is desulfurized prior to entering the reformer with recovery possible as elemental sulfur. There is
essentially no production of NOx as the fuel cell produces electricity through electrochemical reactions and the
combustion in the fuel cell exhaust is at low temperature and in the presence of a high fraction of water. The pro-
duction of CO2, a greenhouse gas is also quite low, approximately 75g carbon/kW/hr, some 3 to 5 times lower than
current coal-fired stations, and nearly 25 percent lower than the gas-fired H-class combined cycle.
The NGT/SOFC hybrid could, of course, be integrated with coal gasification to give a very high performance
coal-based system. This would give additional fuel flexibility to an already attractive power generation system.
3-20
4. REFERENCES
Hogg, S. and Leschziner, M. A., “Computation of Highly Swirling Confined Flow With a Reynolds Stress Turbu-
lence Model,” AIAA J. 27(1): 57 (1989).
Jenny, P., Pope, S. B., Muradoglu, M., and Caughey, D. A., “A Hybrid Algorithm for the Joint PDF Equation of
Turbulent Reactive Flows,” J. Comput. Phys. 166: 252 (2001).
Malecki, R. E., Rhie, C. M., McKinney, R. G., Ouyang, H. Syed, S. A., Colket, M. B., and Madabhushi, R. K.,
“Application of an Advanced CFD-Based Analysis System to the PW6000 Combustor To Optimize Exit Tem-
perature Distribution – PART 1: Description and Validation of the Analysis Tool,” Proc. ASME TURBO
EXPO 2001, 2001-GT-0062 (2001).
Muradoglu, M., Jenny, P., Pope, S. B., and Caughey, D. A., “A Consistent Hybrid Finite-Volume/Particle-Method
for the PDF Equations of Turbulent Reactive Flows,” J. Comput. Physics 154:342 (1999).
Muradoglu, M., Pope, S. B., and Caughey, D. A., “The Hybrid Method for the PDF equations of Turbulent Reac-
tive Flows: Consistency Conditions and Correction Algorithms,” J. Comput. Physics 172: 841-878 (2001).
Muradoglu, M., Pope, S. B., “A Local Time Stepping Algorithm for Solving the PDF Equations for Turbulent
Reacting Flows,” AIAA Journal (submitted 2001).
Pope, S. B., “PDF Methods for Turbulent Reactive Flows,” Prog. Energy Combust. Sci. 11: 119 (1985).
Pope, S. B., “Computationally Efficient Implementation of Combustion Chemistry Using In-Situ Adaptive Tabula-
tion,” Combust. Theory Modeling 1, 41 (1997).
4-1

NTG

Uploaded by

Copyright:

Available Formats

NTG

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NTG

Uploaded by

Copyright:

Available Formats

What is the purpose of the document?

What is the purpose of the document?

What technical goals did the Next Generation Turbine Program target?

What technical goals did the Next Generation Turbine Program target?

TW-0076

NEXT GENERATION TURBINE PROGRAM

TECHNICAL PROGRESS REPORT

Published by Business Development Technical Writing

EXECUTIVE SUMMARY ..................................................................................................................................x

1.1 Introduction ........................................................................................................................................ 1-1

1.3 Economics ......................................................................................................................................... 1-2

1.4 Reliability, Availability, and Maintainability..................................................................................... 1-4

2. RESULTS AND DISCUSSION....................................................................................................................... 2-1

2.1 Component System Descriptions....................................................................................................... 2-1

2.2 Combined Cycle ............................................................................................................................... 2-26

3.1 Market Studies .................................................................................................................................... 3-1

3.2 Public Benefits.................................................................................................................................. 3-10

4. REFERENCES ................................................................................................................................................. 4-1

ADL Arthur D. Little Company

CFD Computational Fluid Dynamics

DLE Dry Low Emissions

EBC Environmental Barrier Coating

HAT Humid Air Turbine

IGCC Integrated Gasification Combined Cycle

LES Large Eddy Simulations

MPPT Multiphase Power and Processing Technologies

NETL National Energy Technology Laboratory

O&M Operating and Maintenance

P&W Pratt & Whitney

RAM Reliability, Availability, and Maintainability

SOFC Solid Oxide Fuel Cells

UTC United Technologies Corporation

• Increase lower heating value net system efficiency by 15 percent or more

• Improve turndown ratios (using a turbine at partial capacity) by 50 percent or more

• Improve service life

• Reduce emissions of carbon and nitrogen oxide gases

• Reduce operations, maintenance, and capital costs by 15 percent or more

• Provide the capability to use multiple fuels.

tems. Program goals include a 15 percent improvement in

efficiency, a 15 percent reduction in operating, mainte-

1.2 OVERALL SYSTEM DESCRIPTION AND ATTRIBUTES

1.3.2 Operating and Maintenance Cost: 225 7EA

FT60 IC PW4090 F-Class Frame Type

gas turbines and one steam turbine), representing a more 200

NGTs vs. FA GT in Peaking

NGTs vs. FA GT in Intermediate

NGTs vs. FX-2 CC in Intermediate

NGTs vs. FX-2 in Base Load

-5% 0% 5% 10% 15% 20%

Annual Operating Hours

2. RESULTS AND DISCUSSION

2.1 COMPONENT SYSTEM DESCRIPTIONS

• No new technology in LPC/HPC for intercool cycle

2.1.1.1 Low-Pressure Compressor

(36) (39) (46) (49)

Blades and Vanes: Low Cost Rotor:

Figure 2-2. Low-Pressure Compressor Module

Cantilevered Fixed Stators

Bolted Rear Hub

Nox — 15% O2, ppmdv

and dφ* = M(φ,...)dt + S(φ)dt