THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St, New York, N.Y

THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St, New York, N.Y. 17 93-GT-146 The Society shall not be responsible for statements or opinions advanced in papers or discussion at meetings of the Society or of its Divisions or Sections, or printed in its publications. Discussion is printed only if the paper is published in an ASME Journal. Papers are available from ASME for 15 months after the meeting. Printed in U.S.A. Copyright 1993 by ASME A NEW HIGH-EFFICIENCY HEAVY-DUTY COMBUSTION TURBINE 71F I. Fukue, S. Aoki, and K. Aoyama S. Umemura Takasago Machinery Works Takasago R&D Center Mitsubishi Heavy Industries, Ltd. Mitsubishi Heavy Industries, Ltd. Takasago, Japan Takasago, Japan A. Merola, M. Noceto, and M. Rosso Fiat Avio S.p.A. Turin, Italy ABSTRACT The 71F is a high temperature 5Hz industrial grade 22 MW size engine based on a scaling of the 51F 15MW class 6Hz machine, and incorporates a higher compressor pressure ratio to increase the thermal efficiency. The prototype engine is under a two-year performance and reliability verification testing program at MHI's Yokohama Plant and was initially fired in June of 1992. This paper describes the 71F design features design changes made from 51 F. The associated performance and reliability verification test program will also be presented. INTRODUCTION The 71F is a 3 RPM heavy-duty combustion turbine designed with a 1.2 scaling factor from the 15MW, 6Hz 51F to serve the 5Hz power generation needs for utility and industrial service. This engine, jointly developed by Mitsubishi Heavy Industries, Ltd., Westinghouse Electric Corporation, and Fiat Avio S. p. A., represents the latest in an evolutionary cycle that continues a long line of large single shaft heavy-duty combustion turbines. The 71F combines the efficient, reliable design concepts of the 51F with recently developed low emission combustion technology. The result is an advanced design, high temperature, efficient, low NOx, more powerful combustion turbine based on time proven reliable design concepts that will satisfy the large combustion turbine power generation needs for the next decade. Currently being targeted for 1992-1994 performance and durability test, it will have an initial simple cycle ISO rating of 221MW with a heat rate of 9,44 BtukWh (9,958 KJkWh) based on LHV at a turbine inlet temperature of 126 C (1533K) on natural gas fuel. In combined cycle applications the heat rate will be better than 6,36 Btu kwh (6,79 KJkWh) based on LHV with a single shaft application. Across the board advances in computer technology have enabled manufacturers to improve analytical procedures in all aspects of design including stress analysis, heat transfer, aerodynamics, fluid mechanics, and structural dynamics. Benefits of these technological advances are shown in Figure 1, where optimal cooling system design allows metal temperature in the 71F to be kept within MW51D5MW71D experience and have been verified in the 51F prototype engine shop test with extensive instrumentation installed in the engine. The first heavy-duty combustion turbine to incorporate advanced cooling technologies was the MF111 with initial commercial operation in August, 1986 at a turbine inlet temperature of 116 C (1433K). Si a ct. 2 14 13 12 1 8 7 6 5 71F51FI OAS TEMP. 71D515 1 2 2 3 3 4 V B V B V B V VANE AND BLADE ROWS Figure 1 71F Metal and Gas Temperatures For some critical components such as combustor baskets, transition pieces, and turbine row 1 and 2 blades and vanes, the 71F engine utilizes the components of the same size or the same basic dimensions as those of the 51F in order to enhance reliability and performance. This design philosophy enables this engine to benefit from the comprehensive 51 F testing program which consisted of air cascades, turbine model tests, full-scale verification of combustor design, and rotor blade vibratory dynamics. The component testing was followed by an instrumented shop test at load and finally by an instrumented field test, which included tests of several design enhancements. In addition, stringent emission regulations in Japan provided increased incentive to develop the dry low NOx Hybrid Combustion system that has operated successfully at Tohoku Electric Power Company Higashi-Niigata Plant since 1984. 4 B Presented at the International Gas Turbine and Aeroengine Congress and Exposition Cincinnati, Ohio May 24-27, 1993 This paper has been accepted for publication in the Transactions of the ASME Discussion of it will be accepted at ASME Headquarters until September 3,1993 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use

ENGINE DESIGN FEATURES Figure 2 illustrates the general configuration of the 71F heavy-duty combustion turbine. Several basic long established design concepts and philosophies are evident and are presented below. Those features unique to the 71F are listed separately. Figure 2 51F Longitudinal Section General Description Two-bearing single shaft construction. Cold end generator drive to minimize the influence of thermal expansion and to eliminate the need for a flexible coupling. Axial flow exhaust system, which has a plant arrangement advantage in combined cycle power plants in meeting the heat recovery requirements. Externally cooled and filtered rotor and blade cooling air to eliminate excessive contaminants that could' block critically intricate cooling passages of the rotor and blades. Tangential exhaust casing struts to maintain bearing housing alignment by rotating as required to accommodate thermal expansion. Variable inlet guide vane (IGV) to provide exhaust temperature control for heat recovery applications and to improve starting characteristics. Compressor diaphragms and dove-tail rooted blades removable with the rotor in place. Three (3) axisymmetric compressor bleeds for turbine cooling with two (2) used to avoid surge during starting. Horizontally split casings to facilitate field maintenance with the rotor in place. Combustors and transitions removable without lifting cylinder covers. Stage I vanes removable without lifting cylinder cover. Turbine rotor with bolted CURVIC coupled discs providing precise alignment and torque carrying features. Fir-tree rooted turbine blades removable on-site with rotor in place. Multiple turbine blade ring concept to provide field service of vanes with the rotor in place and a thermal response independent of the outer casing to prevent blades rubs, minimize clearance, and maximize performance. Turbine ring segments and isolating ring structure to minimize the thermal distortion of the blade rings which support the turbine vanes. 71 F Additional Features High temperature and efficient engine designed based on the proven 51F engine with the scale ratio of 1.2 with the exception of the combustors and turbine row 1 and 2 blade and vane airfoils which are identical to those of the 51F. Advanced hybrid combustion system incorporating low NOx features, consisting of 2 cannular combustors with the same diameter and length as the 51F. The hybrid combustor features a two-stage burner assembly and a bypass valve which directs a portion of the compressor delivery air directly into the transition piece to enhance flame stability during starting and to maintain desired fuelair ratio during loading. The 71 F hybrid system differs from the current one by having the ratio of pilot to main fuel trimmed to reduce pilot burner NOx generation. Twin layer composite structure named "PLATEFIN" and "MTFIN" to provide more efficient cooling on the combustor basket and transition pieces respectively, thus providing more air for the low NOx system. Four (4) stage turbine to maintain low aerodynamic loading even at the increased firing temperature. Cooled stage 1 and 2 vane segment and cooled stage 1 and 2 blades with the basic same dimensions as 51 F engine to utilize the 51F componentshop test results and field experiences. The first two stages of vanes and blades are protected by anti-oxidation coating. The first stage vane shrouds are also protected by thermal barrier coating. The Row 1 vane cooling design utilizes state-of-the-art concepts with three impingement inserts in combination with an array of film cooling exits and a trailing edge pin fin system. The first stage blade is cooled by a combination of convection techniques via multi-pass serpentine passages, pin fin cooling in the trailing edge exit slots, and film cooling including shower head scheme. Cooled stage 3 vane segment and turbine blade to improve reliability. Integral "Z" tip shrouds in third and fourth stage rotor blades for increased structural damping to minimize the potential for flow induced non-synchronous vibration. Damping and sealing pins in the first and second stage rotor blades to increase structural damping and minimize the leakage of cooling flow. Leading Edge Groove (LEG) direct lubricated thrust bearing to reduce the required oil flow and its mechanical loss. Two-element tilting pad journal bearings for load carrying and a fixed upper half bearing to eliminate top pad fluttering concerns and related local babbitt spragging. Compressor blade locking feature that is visibly inspectable. Improved compressor rotor blade root design that has flat contact faces for ease of manufacturing and inspection. Blade rings in compressor section to optimize cylinder to rotor alignment. Compressor rotor design with multiple axial tie bolts which eliminates the shrinkfitted design and main coupling joint to increase rotor dynamic stability margin as well as facilitate fabrication and maintenance of the rotor. Individual disks are positioned radially by bore rabbet joints. 17 stage, highly efficient axial flow compressor with pressure ratio of 16:1, incorporating larger diameter rear stages to help balance spindle thrust and two exit guide vanes to strainghten the flow leaving the compressor. Turbine flow path design utilizing fully three-dimensional flow analysis. COMPONENT VERIFICATION For some critical components, the 71F engine utilizes components of the same size or same basic dimensions as the 51F. This design philosophy enables the engine to benefit from the comprehensive 51F test program and operating experience. As an integral part of the 71 F engine development program, performance and durability testing including rotating blade vibration tests, a two-year performance and durability test program for performance measurement in various ambient conditions, and durability and emissions verifications in long-term daily startstop operation. In the test at MHI's newly constructed power station operating on natural gas in Yokohama, the generator was coupled with the 71F engine in the same manner as in the simple cycle unit with the downstream high temperature SCR (selective catalytic removal) system as shown in Figure 3. The generated electrical output is delivered to 5Hz utility grid. About 13 special measurement 2 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use

points were applied to the engine during the integrated testing program. CH I cr. Blade to sensor 1 2 CH 2 Reference pos tron detector ri 3 CH. 3 f-r m-11,...e, m CH. m M1 Light source Data mernor Data analyzer Figure 4 Blade Vibratory Stress Measuring System Figure 3 71F Gas Turbine Plant in Yokohama Rotating Blade Vibration Test A rotating blade vibration test was performed to verify the vibration characteristics of selected rotating compressor and turbine blades. The first and second stage compressor blades and all four turbine stage blades were tested to verify natural frequencies. Damping characteristics were also measured for the turbine 1st and 2nd stage blades with seal pins and also the 3rd and 4th stage shrouded blades of the turbine. These damping characteristics will be used to obtain turbine blade dynamic response via a cyclic symmetry computer routine. During the test, performed in a vacuum room, the fully bladed rotor was driven by an electric motor up to 11% of rated speed. The frequencies and amplitude of compressor blade vibration were measured using a non-contact optical fiber monitoring system. This system measures the movement of each individual blade tip by monitoring the position of each blade tip from several different locations circumferentially and analyzing this information through a synchronizing process. The schematic of this measuring system is shown in Figure 4. In the turbine, strain gauges were applied directly to the blades of each stage and the measured strain signals were transmitted to a data acquisition system through a telemetry system. After analyzing all measured data, it was confirmed that no further tuning was necessary for the blades. Engine Verification Test In order to verify the performance and design characteristics of the 71 F, a two-year test program for performance, durability and emissions was started in June, 1992. Figure 5 presents an overall general arrangement of the Mitsubishi Power Station, showing the major components: gas turbine, generator, exciter, control and EXHAUST STACK EXHAUST STAC COOLING AIR COOLER 71F \ GAS TURBINE SILENCER EXHAUST DUCT -=.Rr: 7; <_.ari. _, &Mei ARM11 I ' 1111111.11 611.,111)11173n p. uptibra WAI-411 11"11 I 1,11171 1 ";11141I "?1111 ta. ' 11111' 11 A A -- COOLING WATER PUMP GENERATOR 1.1,1 11.1111 1=1.1., SILENCER irairgli AUXILIARY PACKAGE INTAKE DUCT INLET FILTER C CONTROL ROOM - - Figure 5 Plant Layout in Yokohama 3 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use

PERFORMANCE AIR FLOW INLET TEMP. & PRESS EXHAUST TEMP. & PRESS FUEL FLOW GENERATOR OUTPUT COMPRESSOR SURGE MARGIN STAGE EFF. TURBINE STAGE EFF. DIFFUSER EFF. I MEASUREMENT ITEMS METAL TEMP. STRESSVIBRATION COMBUSTOR BASKET TRANSITION PIECE TURB. ROW1 BLADE TURB. ROW1-4 VANE ' BEARING METAL OUTER CASING INNER CASING EXHAUST CYLINDER COMPRESSOR BLADE ffi COMPRESSOR VANE COMBUSTOR BASKET TRANSITION PIECE C!' TURBINE BLADE ROTOR VIB. CASING VIB. ROTOR TORSIONAL VIB. OTHERS ii4 COOLING AIR NETWORK FLOW TEMPERATURE PRESSURE g THRUST LOAD EXHAUST EMISSION ROTORCASING EXPANSION NOISE LUBE OIL TEMP. all m ID op Fla 1941VIIN RION itflv: - = k, 3,(1.4.1a a vile& 1911A1111 special instrumentation room, cooling air cooler, gas compressor, SCR, and intake and exhaust stacks. About 1,3 special measurement points were applied to the engine for the confirmation of the following: Individual compressor and turbine performance as well as overall gas turbine performance. Parameters included air flow, power output, heat rate and exhaust temperature. Compressor inlet air flow over the entire IGV range. The compressor air supply duct system employed a bell-mouth inlet to measure air flow precisely. Engine starting and acceleration characteristics including lightoff, rotor vibration, and rotating stall. Mechanical operation of the engine from starting to overspeed including rotor vibration characteristics. Mechanical and thermal performance of the engine over its entire operating range. Reliability of the engine by measurement of gas and metal temperature, pressure, vibratory stresses, etc. The hot parts' metal temperatures will be confirmed to be below allowables at the associated 126 C (1533K) rotor inlet temperature. Monitored components include combustor basket, transitions, turbine vanes, blades, and turbine ring segment. Vibratory stresses of the compressor and turbine blades, bearing temperatures, casing temperature, and disc cavity temperatures are monitored continuously during the test. In addition, axial and radial growths are monitored at strategic locations in the engine to verify design calculations. Emission characteristics of the engine. Emissions testing includes sampling for nitrogen oxides (NOx), nitric oxide (NO), volatile organic compounds (VOC), carbon monoxide (CO), unburned hydrocarbons (UHC), carbon dioxide (CO 2), and oxygen ( 2). Monitoring System during Engine Operation In addition to the supervisory instrumentation, the engine is extensively instrumented to measure thermodynamic performance parameters, metal temperatures, static and vibratory strains, vibration characteristics, displacements, and other parameters as shown in Figure 6. Dynamic strain gauges will be installed on the turbine blades to verify dynamic responses. The signals from the rotating sensors are transmitted by a telemetry system. Clearance measurement ;IF 'V 117)21 4--aik - -11.1 16 t-_-e -Alga4t4 IAILWRIntlir e ' 11111 11111V-1_ Figure 6 71F Prototype Special Measurement system using proximity probes allow stator to rotor radial and axial displacement measurements during transients. Through the use of an infrared pyrometer, it will be possible to obtain the temperature distribution on each turbine blade of the first stage under operating conditions. Data acquisition equipment is installed to record the special engineering test data. This equipment includes tape recorders, spectrum analyzers, plotters and chart recorders. Data critical to the continued operation of the engine, such as metal temperatures of hot parts and cavity temperatures, are monitored on computer displays together with associated alarm limits. Turbine inlet temperature of the operating engine can be calculated from measured data using a heat balance calculation program. TEST RESULTS The 71F performance and durability testing is a two-year program which started in June, 1992. Summarized below are the test results measured at the initial stage of this program. Starting and Acceleration This plant is started by operating the generator as a motor via a static frequency converter. The starting schedule includes a 5-minute exhaust stack purge, ignition at 55-6rpm, disengagement of the static frequency converter at 2rpm, an IGV position change at 2rpm and bleed valve closure at 2755rpm. The elapsed time from pushing starting button to synchronizing speed is under 18 minutes and time to full load is under 37 minutes as shown in Figure 7. At ignition speed, two combustors located at the bottom of the engine are ignited and cross-flame tubes produce complete ignition. Compressor rotating stall was cleared at less than 17rpm. Throughout the starting sequence, compressor operating characteristics were stable. Overall Performance Measured performance data were analyzed and corrected from shop test conditions to standard ambient conditions. The results show that the measured power output and inlet air flow exceeded the predicted values. In addition, the heat rate is slightly better than the predicted value. Since performance will be measured 4 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use

for the various ambient conditions during the two-year test program. new performance ratings will be established after completion of the test program. temperature at design conditions. Evaluation of hot parts data is continuing and more detailed turbine blade and vane temperatures will be verified in the next stage in the test program. Start - Ignition : 9mln. Start rated speed :113min. Start - lull load : 37mln., - Generator output Sr 5 8 O. ALLOWABLE LIMITS TO Location - - 97%Haighl n 74% I -2 o 5% O 27% -3 -c 3% O a 122-4 (.5-5 O Figure 7 Starting Characteristics 3 35 4 45 5 15 GENERATOR OUTPUT [MIN] Figure 9 Turbine Row 1 Vane Metal Temperature Table 1 Performance Results EXPECTED MEASURED POWER, NET KW 224,81 +3.8% HEAT RATE, KJKW-HR 9,958 1.9%, BTUKW-HR 9,44 1.9% AIR FLOW, KGSEC 655 +1.%, LBSSEC 1,44 +1.% EXHAUST TEMPERATURE, C 554 3.C,K 827 3.K Emission Nitrogen oxides (NOx), carbon monoxide (CO), unburned hydrocarbons (UHC), volatile organic compound (VOC), and other emissions were measured. Since premixed type combustors are used, the setting schedules of inlet guide vanes and bypass valves affect the combustion performance. Figure 8 shows the dynamic pressure fluctuation measured inside the combustor. Stable combustion was assured throughout engine start up and load operation. Low NOx emission level was measured not only at full load condition, but also at lower load conditions including idle condition. At rated load (15MW) of this plant, measured emission levels were by far lower than the target level of 5ppm at dry conditions. 5 ALLOWABLE LIMITS A COMBUSTOR NO. P-5 X P-1 A P-13 A A O P-16 )11IK Figure 1 Combustor Metal Temperature Vibratory Stresses Figure 11 presents vibratory stresses as measured during the test along with corresponding design allowables on the Goodman diagram. Frequency and damping factors were also verified during the test. Vibratory stresses of other stationary components such as a combustor transition, tangential strut shield cover, exhaust diffuser support, etc., were measured. As shown in Figure 12, all measured stresses were within the design allowables. Figure 13 summarizes rotor dynamic responses during start up. Vibration levels were low during acceleration to running speed and the associated phase angles during rated speed operation were virtually constant. DX AIL Lr 5 15 LOAD (MW) Figure 8 Combustor Pressure Fluctuation 5 5 STATIC STRESS 1% OF YIELD] STATIC STRESS (16 OF YIELD] Hot Parts Temperature Figure 9 presents metal temperatures measured along the span of a row 1 vane leading edge. The temperatures were obtained from thermocouples which were mounted on the surface of the vane airfoils. Figure 1 presents the combustor system metal temperatures. It was confirmed that the mesured hot parts temperatures were within the acceptable range and this also verified expected Row 2 Blade Row 4 Blade Figure 11 Compressor Blade Vibratory Stress 5 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use

COMPRESSOR ROW 2 VANE ROW 6 VANE COMBUSTOR BASKET TRANSITION U-SUPPORT FLEXIBLE SUPPORT 771 EXHAUST CASINO DIFFUSER SUPPORT STRUT SHIELD COVER INNER CASINO SUPPORT 1 2 7 4 5 GO 7 SO VIBRATORY STRESS NORMALIZED TO ALLOWABLE LIMIT [4] Figure 12 Summary of Vibratory Stress 1 TURBINE BEARING Cooling Flow Circuit The measurement results established the correlations between the various engine cavity temperatures and the supervisory instrumentation installed in all engines to assure that specific temperature limits are never exceeded during normal operation. The individual cooling circuit flows were measured via orifice plates installed in the external piping while important engine cavity temperatures were monitored simultaneously with supervisory and special test thermocouples. FUTURE TEST PLAN The two-year test program of the first 71 F engine started at the beginning of June, 1992. Initially, the engine was operated in an introductory rating mode, i.e., the rotor inlet temperature was reduced. This is the only prudent way to introduce a reliable engine to the market which features a step change in firing temperature technology. Mechanical technologies necessary to achieve significant increase in turbine firing temperature should be verified via operating experience prior to operation at the rated rotor inlet temperature of 126 C (1533K). This will result in a more reliable product when operated at rated conditions. SPEED (RPM) Figure 13 Rotor Vibration Clearances Figure 14 presents axial clearance evaluations for both cold and hot start conditions. The hot start measurements were taken shortly after a scheduled turbine shutdown from full load, which analytically is the severest condition for axial clearance verification. Tip clearances of compressor 14th stage and turbine 2nd stage were also measured. Since the 2nd stage turbine tip clearance was slightly larger than expected, turbine performance will be increased by reducing this clearance. 4 SUMMARY The 71F prototype engine performance and durability testing has been continuing successfully since June, 1992. Overall performance is better than anticipated and NOx emission levels are far below the target. The operation of the generator as a motor via a static frequency converter was verified to be adequate during the starting and acceleration testing. Mechanical characteristics such as vibratory stresses of blades and diaphragms, thrust force, axialradial clearances, and bearing and turbine blade and vane metal temperatures were confirmed to be within the design limits. The two-year test program includes more detailed data measurements at various operating conditions. Recently developed low NOx combustor and other associated design enhancements will also be verified in this program. 7e" 4 5 3 2 T, E 2 4 6 BO 12 LOAD SPEED ' I 14 16 18 2 C 28 3 32 48 TIME (MIN.) SOO 52 REFERENCES 1. Entenmann, D. T., Fukue, I., et al., "Shop Test of the 51F A 15MW Combustion Turbine," ASME Paper No.9-GT-362. 2. Scalzo, A. J., Mori, Y., et al., "A New 15MW High Efficiency Heavy-Duty Combustion Turbine," ASME Paper No.88-GT- 162. 3. Scalzo, A. J., Fukue I., et al., "Operating Experience Complements New Technology in Design of Advanced Combustion Turbine," ASME Paper No.88-JPGCGT-1. 4. Entenmann, D. T., Tsukagoshi, K., et al., "51F Development Update," ASME Paper No.92-GT-237. cmw 2 1 w C.) g,- a. cn E 5 E cc - g cc R Q 2 to cc 1-- cc E,-.. 2 4... " 6 6 12 14 16 18 2 28 3 32 48 5 52 TIME (MIN.) Figure 14 RotorCasing Relative Axial Displacement 6 Downloaded From: http:proceedings.asmedigitalcollection.asme.org on 58218 Terms of Use: http:www.asme.orgabout-asmeterms-of-use